Annotation of imach/src/imach.c, revision 1.259
1.259 ! brouard 1: /* $Id: imach.c,v 1.258 2017/04/03 10:17:47 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.259 ! brouard 4: Revision 1.258 2017/04/03 10:17:47 brouard
! 5: Summary: Version 0.99r12
! 6:
! 7: Some cleanings, conformed with updated documentation.
! 8:
1.258 brouard 9: Revision 1.257 2017/03/29 16:53:30 brouard
10: Summary: Temp
11:
1.257 brouard 12: Revision 1.256 2017/03/27 05:50:23 brouard
13: Summary: Temporary
14:
1.256 brouard 15: Revision 1.255 2017/03/08 16:02:28 brouard
16: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
17:
1.255 brouard 18: Revision 1.254 2017/03/08 07:13:00 brouard
19: Summary: Fixing data parameter line
20:
1.254 brouard 21: Revision 1.253 2016/12/15 11:59:41 brouard
22: Summary: 0.99 in progress
23:
1.253 brouard 24: Revision 1.252 2016/09/15 21:15:37 brouard
25: *** empty log message ***
26:
1.252 brouard 27: Revision 1.251 2016/09/15 15:01:13 brouard
28: Summary: not working
29:
1.251 brouard 30: Revision 1.250 2016/09/08 16:07:27 brouard
31: Summary: continue
32:
1.250 brouard 33: Revision 1.249 2016/09/07 17:14:18 brouard
34: Summary: Starting values from frequencies
35:
1.249 brouard 36: Revision 1.248 2016/09/07 14:10:18 brouard
37: *** empty log message ***
38:
1.248 brouard 39: Revision 1.247 2016/09/02 11:11:21 brouard
40: *** empty log message ***
41:
1.247 brouard 42: Revision 1.246 2016/09/02 08:49:22 brouard
43: *** empty log message ***
44:
1.246 brouard 45: Revision 1.245 2016/09/02 07:25:01 brouard
46: *** empty log message ***
47:
1.245 brouard 48: Revision 1.244 2016/09/02 07:17:34 brouard
49: *** empty log message ***
50:
1.244 brouard 51: Revision 1.243 2016/09/02 06:45:35 brouard
52: *** empty log message ***
53:
1.243 brouard 54: Revision 1.242 2016/08/30 15:01:20 brouard
55: Summary: Fixing a lots
56:
1.242 brouard 57: Revision 1.241 2016/08/29 17:17:25 brouard
58: Summary: gnuplot problem in Back projection to fix
59:
1.241 brouard 60: Revision 1.240 2016/08/29 07:53:18 brouard
61: Summary: Better
62:
1.240 brouard 63: Revision 1.239 2016/08/26 15:51:03 brouard
64: Summary: Improvement in Powell output in order to copy and paste
65:
66: Author:
67:
1.239 brouard 68: Revision 1.238 2016/08/26 14:23:35 brouard
69: Summary: Starting tests of 0.99
70:
1.238 brouard 71: Revision 1.237 2016/08/26 09:20:19 brouard
72: Summary: to valgrind
73:
1.237 brouard 74: Revision 1.236 2016/08/25 10:50:18 brouard
75: *** empty log message ***
76:
1.236 brouard 77: Revision 1.235 2016/08/25 06:59:23 brouard
78: *** empty log message ***
79:
1.235 brouard 80: Revision 1.234 2016/08/23 16:51:20 brouard
81: *** empty log message ***
82:
1.234 brouard 83: Revision 1.233 2016/08/23 07:40:50 brouard
84: Summary: not working
85:
1.233 brouard 86: Revision 1.232 2016/08/22 14:20:21 brouard
87: Summary: not working
88:
1.232 brouard 89: Revision 1.231 2016/08/22 07:17:15 brouard
90: Summary: not working
91:
1.231 brouard 92: Revision 1.230 2016/08/22 06:55:53 brouard
93: Summary: Not working
94:
1.230 brouard 95: Revision 1.229 2016/07/23 09:45:53 brouard
96: Summary: Completing for func too
97:
1.229 brouard 98: Revision 1.228 2016/07/22 17:45:30 brouard
99: Summary: Fixing some arrays, still debugging
100:
1.227 brouard 101: Revision 1.226 2016/07/12 18:42:34 brouard
102: Summary: temp
103:
1.226 brouard 104: Revision 1.225 2016/07/12 08:40:03 brouard
105: Summary: saving but not running
106:
1.225 brouard 107: Revision 1.224 2016/07/01 13:16:01 brouard
108: Summary: Fixes
109:
1.224 brouard 110: Revision 1.223 2016/02/19 09:23:35 brouard
111: Summary: temporary
112:
1.223 brouard 113: Revision 1.222 2016/02/17 08:14:50 brouard
114: Summary: Probably last 0.98 stable version 0.98r6
115:
1.222 brouard 116: Revision 1.221 2016/02/15 23:35:36 brouard
117: Summary: minor bug
118:
1.220 brouard 119: Revision 1.219 2016/02/15 00:48:12 brouard
120: *** empty log message ***
121:
1.219 brouard 122: Revision 1.218 2016/02/12 11:29:23 brouard
123: Summary: 0.99 Back projections
124:
1.218 brouard 125: Revision 1.217 2015/12/23 17:18:31 brouard
126: Summary: Experimental backcast
127:
1.217 brouard 128: Revision 1.216 2015/12/18 17:32:11 brouard
129: Summary: 0.98r4 Warning and status=-2
130:
131: Version 0.98r4 is now:
132: - displaying an error when status is -1, date of interview unknown and date of death known;
133: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
134: Older changes concerning s=-2, dating from 2005 have been supersed.
135:
1.216 brouard 136: Revision 1.215 2015/12/16 08:52:24 brouard
137: Summary: 0.98r4 working
138:
1.215 brouard 139: Revision 1.214 2015/12/16 06:57:54 brouard
140: Summary: temporary not working
141:
1.214 brouard 142: Revision 1.213 2015/12/11 18:22:17 brouard
143: Summary: 0.98r4
144:
1.213 brouard 145: Revision 1.212 2015/11/21 12:47:24 brouard
146: Summary: minor typo
147:
1.212 brouard 148: Revision 1.211 2015/11/21 12:41:11 brouard
149: Summary: 0.98r3 with some graph of projected cross-sectional
150:
151: Author: Nicolas Brouard
152:
1.211 brouard 153: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 154: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 155: Summary: Adding ftolpl parameter
156: Author: N Brouard
157:
158: We had difficulties to get smoothed confidence intervals. It was due
159: to the period prevalence which wasn't computed accurately. The inner
160: parameter ftolpl is now an outer parameter of the .imach parameter
161: file after estepm. If ftolpl is small 1.e-4 and estepm too,
162: computation are long.
163:
1.209 brouard 164: Revision 1.208 2015/11/17 14:31:57 brouard
165: Summary: temporary
166:
1.208 brouard 167: Revision 1.207 2015/10/27 17:36:57 brouard
168: *** empty log message ***
169:
1.207 brouard 170: Revision 1.206 2015/10/24 07:14:11 brouard
171: *** empty log message ***
172:
1.206 brouard 173: Revision 1.205 2015/10/23 15:50:53 brouard
174: Summary: 0.98r3 some clarification for graphs on likelihood contributions
175:
1.205 brouard 176: Revision 1.204 2015/10/01 16:20:26 brouard
177: Summary: Some new graphs of contribution to likelihood
178:
1.204 brouard 179: Revision 1.203 2015/09/30 17:45:14 brouard
180: Summary: looking at better estimation of the hessian
181:
182: Also a better criteria for convergence to the period prevalence And
183: therefore adding the number of years needed to converge. (The
184: prevalence in any alive state shold sum to one
185:
1.203 brouard 186: Revision 1.202 2015/09/22 19:45:16 brouard
187: Summary: Adding some overall graph on contribution to likelihood. Might change
188:
1.202 brouard 189: Revision 1.201 2015/09/15 17:34:58 brouard
190: Summary: 0.98r0
191:
192: - Some new graphs like suvival functions
193: - Some bugs fixed like model=1+age+V2.
194:
1.201 brouard 195: Revision 1.200 2015/09/09 16:53:55 brouard
196: Summary: Big bug thanks to Flavia
197:
198: Even model=1+age+V2. did not work anymore
199:
1.200 brouard 200: Revision 1.199 2015/09/07 14:09:23 brouard
201: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
202:
1.199 brouard 203: Revision 1.198 2015/09/03 07:14:39 brouard
204: Summary: 0.98q5 Flavia
205:
1.198 brouard 206: Revision 1.197 2015/09/01 18:24:39 brouard
207: *** empty log message ***
208:
1.197 brouard 209: Revision 1.196 2015/08/18 23:17:52 brouard
210: Summary: 0.98q5
211:
1.196 brouard 212: Revision 1.195 2015/08/18 16:28:39 brouard
213: Summary: Adding a hack for testing purpose
214:
215: After reading the title, ftol and model lines, if the comment line has
216: a q, starting with #q, the answer at the end of the run is quit. It
217: permits to run test files in batch with ctest. The former workaround was
218: $ echo q | imach foo.imach
219:
1.195 brouard 220: Revision 1.194 2015/08/18 13:32:00 brouard
221: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
222:
1.194 brouard 223: Revision 1.193 2015/08/04 07:17:42 brouard
224: Summary: 0.98q4
225:
1.193 brouard 226: Revision 1.192 2015/07/16 16:49:02 brouard
227: Summary: Fixing some outputs
228:
1.192 brouard 229: Revision 1.191 2015/07/14 10:00:33 brouard
230: Summary: Some fixes
231:
1.191 brouard 232: Revision 1.190 2015/05/05 08:51:13 brouard
233: Summary: Adding digits in output parameters (7 digits instead of 6)
234:
235: Fix 1+age+.
236:
1.190 brouard 237: Revision 1.189 2015/04/30 14:45:16 brouard
238: Summary: 0.98q2
239:
1.189 brouard 240: Revision 1.188 2015/04/30 08:27:53 brouard
241: *** empty log message ***
242:
1.188 brouard 243: Revision 1.187 2015/04/29 09:11:15 brouard
244: *** empty log message ***
245:
1.187 brouard 246: Revision 1.186 2015/04/23 12:01:52 brouard
247: Summary: V1*age is working now, version 0.98q1
248:
249: Some codes had been disabled in order to simplify and Vn*age was
250: working in the optimization phase, ie, giving correct MLE parameters,
251: but, as usual, outputs were not correct and program core dumped.
252:
1.186 brouard 253: Revision 1.185 2015/03/11 13:26:42 brouard
254: Summary: Inclusion of compile and links command line for Intel Compiler
255:
1.185 brouard 256: Revision 1.184 2015/03/11 11:52:39 brouard
257: Summary: Back from Windows 8. Intel Compiler
258:
1.184 brouard 259: Revision 1.183 2015/03/10 20:34:32 brouard
260: Summary: 0.98q0, trying with directest, mnbrak fixed
261:
262: We use directest instead of original Powell test; probably no
263: incidence on the results, but better justifications;
264: We fixed Numerical Recipes mnbrak routine which was wrong and gave
265: wrong results.
266:
1.183 brouard 267: Revision 1.182 2015/02/12 08:19:57 brouard
268: Summary: Trying to keep directest which seems simpler and more general
269: Author: Nicolas Brouard
270:
1.182 brouard 271: Revision 1.181 2015/02/11 23:22:24 brouard
272: Summary: Comments on Powell added
273:
274: Author:
275:
1.181 brouard 276: Revision 1.180 2015/02/11 17:33:45 brouard
277: Summary: Finishing move from main to function (hpijx and prevalence_limit)
278:
1.180 brouard 279: Revision 1.179 2015/01/04 09:57:06 brouard
280: Summary: back to OS/X
281:
1.179 brouard 282: Revision 1.178 2015/01/04 09:35:48 brouard
283: *** empty log message ***
284:
1.178 brouard 285: Revision 1.177 2015/01/03 18:40:56 brouard
286: Summary: Still testing ilc32 on OSX
287:
1.177 brouard 288: Revision 1.176 2015/01/03 16:45:04 brouard
289: *** empty log message ***
290:
1.176 brouard 291: Revision 1.175 2015/01/03 16:33:42 brouard
292: *** empty log message ***
293:
1.175 brouard 294: Revision 1.174 2015/01/03 16:15:49 brouard
295: Summary: Still in cross-compilation
296:
1.174 brouard 297: Revision 1.173 2015/01/03 12:06:26 brouard
298: Summary: trying to detect cross-compilation
299:
1.173 brouard 300: Revision 1.172 2014/12/27 12:07:47 brouard
301: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
302:
1.172 brouard 303: Revision 1.171 2014/12/23 13:26:59 brouard
304: Summary: Back from Visual C
305:
306: Still problem with utsname.h on Windows
307:
1.171 brouard 308: Revision 1.170 2014/12/23 11:17:12 brouard
309: Summary: Cleaning some \%% back to %%
310:
311: The escape was mandatory for a specific compiler (which one?), but too many warnings.
312:
1.170 brouard 313: Revision 1.169 2014/12/22 23:08:31 brouard
314: Summary: 0.98p
315:
316: Outputs some informations on compiler used, OS etc. Testing on different platforms.
317:
1.169 brouard 318: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 319: Summary: update
1.169 brouard 320:
1.168 brouard 321: Revision 1.167 2014/12/22 13:50:56 brouard
322: Summary: Testing uname and compiler version and if compiled 32 or 64
323:
324: Testing on Linux 64
325:
1.167 brouard 326: Revision 1.166 2014/12/22 11:40:47 brouard
327: *** empty log message ***
328:
1.166 brouard 329: Revision 1.165 2014/12/16 11:20:36 brouard
330: Summary: After compiling on Visual C
331:
332: * imach.c (Module): Merging 1.61 to 1.162
333:
1.165 brouard 334: Revision 1.164 2014/12/16 10:52:11 brouard
335: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
336:
337: * imach.c (Module): Merging 1.61 to 1.162
338:
1.164 brouard 339: Revision 1.163 2014/12/16 10:30:11 brouard
340: * imach.c (Module): Merging 1.61 to 1.162
341:
1.163 brouard 342: Revision 1.162 2014/09/25 11:43:39 brouard
343: Summary: temporary backup 0.99!
344:
1.162 brouard 345: Revision 1.1 2014/09/16 11:06:58 brouard
346: Summary: With some code (wrong) for nlopt
347:
348: Author:
349:
350: Revision 1.161 2014/09/15 20:41:41 brouard
351: Summary: Problem with macro SQR on Intel compiler
352:
1.161 brouard 353: Revision 1.160 2014/09/02 09:24:05 brouard
354: *** empty log message ***
355:
1.160 brouard 356: Revision 1.159 2014/09/01 10:34:10 brouard
357: Summary: WIN32
358: Author: Brouard
359:
1.159 brouard 360: Revision 1.158 2014/08/27 17:11:51 brouard
361: *** empty log message ***
362:
1.158 brouard 363: Revision 1.157 2014/08/27 16:26:55 brouard
364: Summary: Preparing windows Visual studio version
365: Author: Brouard
366:
367: In order to compile on Visual studio, time.h is now correct and time_t
368: and tm struct should be used. difftime should be used but sometimes I
369: just make the differences in raw time format (time(&now).
370: Trying to suppress #ifdef LINUX
371: Add xdg-open for __linux in order to open default browser.
372:
1.157 brouard 373: Revision 1.156 2014/08/25 20:10:10 brouard
374: *** empty log message ***
375:
1.156 brouard 376: Revision 1.155 2014/08/25 18:32:34 brouard
377: Summary: New compile, minor changes
378: Author: Brouard
379:
1.155 brouard 380: Revision 1.154 2014/06/20 17:32:08 brouard
381: Summary: Outputs now all graphs of convergence to period prevalence
382:
1.154 brouard 383: Revision 1.153 2014/06/20 16:45:46 brouard
384: Summary: If 3 live state, convergence to period prevalence on same graph
385: Author: Brouard
386:
1.153 brouard 387: Revision 1.152 2014/06/18 17:54:09 brouard
388: Summary: open browser, use gnuplot on same dir than imach if not found in the path
389:
1.152 brouard 390: Revision 1.151 2014/06/18 16:43:30 brouard
391: *** empty log message ***
392:
1.151 brouard 393: Revision 1.150 2014/06/18 16:42:35 brouard
394: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
395: Author: brouard
396:
1.150 brouard 397: Revision 1.149 2014/06/18 15:51:14 brouard
398: Summary: Some fixes in parameter files errors
399: Author: Nicolas Brouard
400:
1.149 brouard 401: Revision 1.148 2014/06/17 17:38:48 brouard
402: Summary: Nothing new
403: Author: Brouard
404:
405: Just a new packaging for OS/X version 0.98nS
406:
1.148 brouard 407: Revision 1.147 2014/06/16 10:33:11 brouard
408: *** empty log message ***
409:
1.147 brouard 410: Revision 1.146 2014/06/16 10:20:28 brouard
411: Summary: Merge
412: Author: Brouard
413:
414: Merge, before building revised version.
415:
1.146 brouard 416: Revision 1.145 2014/06/10 21:23:15 brouard
417: Summary: Debugging with valgrind
418: Author: Nicolas Brouard
419:
420: Lot of changes in order to output the results with some covariates
421: After the Edimburgh REVES conference 2014, it seems mandatory to
422: improve the code.
423: No more memory valgrind error but a lot has to be done in order to
424: continue the work of splitting the code into subroutines.
425: Also, decodemodel has been improved. Tricode is still not
426: optimal. nbcode should be improved. Documentation has been added in
427: the source code.
428:
1.144 brouard 429: Revision 1.143 2014/01/26 09:45:38 brouard
430: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
431:
432: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
433: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
434:
1.143 brouard 435: Revision 1.142 2014/01/26 03:57:36 brouard
436: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
437:
438: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
439:
1.142 brouard 440: Revision 1.141 2014/01/26 02:42:01 brouard
441: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
442:
1.141 brouard 443: Revision 1.140 2011/09/02 10:37:54 brouard
444: Summary: times.h is ok with mingw32 now.
445:
1.140 brouard 446: Revision 1.139 2010/06/14 07:50:17 brouard
447: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
448: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
449:
1.139 brouard 450: Revision 1.138 2010/04/30 18:19:40 brouard
451: *** empty log message ***
452:
1.138 brouard 453: Revision 1.137 2010/04/29 18:11:38 brouard
454: (Module): Checking covariates for more complex models
455: than V1+V2. A lot of change to be done. Unstable.
456:
1.137 brouard 457: Revision 1.136 2010/04/26 20:30:53 brouard
458: (Module): merging some libgsl code. Fixing computation
459: of likelione (using inter/intrapolation if mle = 0) in order to
460: get same likelihood as if mle=1.
461: Some cleaning of code and comments added.
462:
1.136 brouard 463: Revision 1.135 2009/10/29 15:33:14 brouard
464: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
465:
1.135 brouard 466: Revision 1.134 2009/10/29 13:18:53 brouard
467: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
468:
1.134 brouard 469: Revision 1.133 2009/07/06 10:21:25 brouard
470: just nforces
471:
1.133 brouard 472: Revision 1.132 2009/07/06 08:22:05 brouard
473: Many tings
474:
1.132 brouard 475: Revision 1.131 2009/06/20 16:22:47 brouard
476: Some dimensions resccaled
477:
1.131 brouard 478: Revision 1.130 2009/05/26 06:44:34 brouard
479: (Module): Max Covariate is now set to 20 instead of 8. A
480: lot of cleaning with variables initialized to 0. Trying to make
481: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
482:
1.130 brouard 483: Revision 1.129 2007/08/31 13:49:27 lievre
484: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
485:
1.129 lievre 486: Revision 1.128 2006/06/30 13:02:05 brouard
487: (Module): Clarifications on computing e.j
488:
1.128 brouard 489: Revision 1.127 2006/04/28 18:11:50 brouard
490: (Module): Yes the sum of survivors was wrong since
491: imach-114 because nhstepm was no more computed in the age
492: loop. Now we define nhstepma in the age loop.
493: (Module): In order to speed up (in case of numerous covariates) we
494: compute health expectancies (without variances) in a first step
495: and then all the health expectancies with variances or standard
496: deviation (needs data from the Hessian matrices) which slows the
497: computation.
498: In the future we should be able to stop the program is only health
499: expectancies and graph are needed without standard deviations.
500:
1.127 brouard 501: Revision 1.126 2006/04/28 17:23:28 brouard
502: (Module): Yes the sum of survivors was wrong since
503: imach-114 because nhstepm was no more computed in the age
504: loop. Now we define nhstepma in the age loop.
505: Version 0.98h
506:
1.126 brouard 507: Revision 1.125 2006/04/04 15:20:31 lievre
508: Errors in calculation of health expectancies. Age was not initialized.
509: Forecasting file added.
510:
511: Revision 1.124 2006/03/22 17:13:53 lievre
512: Parameters are printed with %lf instead of %f (more numbers after the comma).
513: The log-likelihood is printed in the log file
514:
515: Revision 1.123 2006/03/20 10:52:43 brouard
516: * imach.c (Module): <title> changed, corresponds to .htm file
517: name. <head> headers where missing.
518:
519: * imach.c (Module): Weights can have a decimal point as for
520: English (a comma might work with a correct LC_NUMERIC environment,
521: otherwise the weight is truncated).
522: Modification of warning when the covariates values are not 0 or
523: 1.
524: Version 0.98g
525:
526: Revision 1.122 2006/03/20 09:45:41 brouard
527: (Module): Weights can have a decimal point as for
528: English (a comma might work with a correct LC_NUMERIC environment,
529: otherwise the weight is truncated).
530: Modification of warning when the covariates values are not 0 or
531: 1.
532: Version 0.98g
533:
534: Revision 1.121 2006/03/16 17:45:01 lievre
535: * imach.c (Module): Comments concerning covariates added
536:
537: * imach.c (Module): refinements in the computation of lli if
538: status=-2 in order to have more reliable computation if stepm is
539: not 1 month. Version 0.98f
540:
541: Revision 1.120 2006/03/16 15:10:38 lievre
542: (Module): refinements in the computation of lli if
543: status=-2 in order to have more reliable computation if stepm is
544: not 1 month. Version 0.98f
545:
546: Revision 1.119 2006/03/15 17:42:26 brouard
547: (Module): Bug if status = -2, the loglikelihood was
548: computed as likelihood omitting the logarithm. Version O.98e
549:
550: Revision 1.118 2006/03/14 18:20:07 brouard
551: (Module): varevsij Comments added explaining the second
552: table of variances if popbased=1 .
553: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
554: (Module): Function pstamp added
555: (Module): Version 0.98d
556:
557: Revision 1.117 2006/03/14 17:16:22 brouard
558: (Module): varevsij Comments added explaining the second
559: table of variances if popbased=1 .
560: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
561: (Module): Function pstamp added
562: (Module): Version 0.98d
563:
564: Revision 1.116 2006/03/06 10:29:27 brouard
565: (Module): Variance-covariance wrong links and
566: varian-covariance of ej. is needed (Saito).
567:
568: Revision 1.115 2006/02/27 12:17:45 brouard
569: (Module): One freematrix added in mlikeli! 0.98c
570:
571: Revision 1.114 2006/02/26 12:57:58 brouard
572: (Module): Some improvements in processing parameter
573: filename with strsep.
574:
575: Revision 1.113 2006/02/24 14:20:24 brouard
576: (Module): Memory leaks checks with valgrind and:
577: datafile was not closed, some imatrix were not freed and on matrix
578: allocation too.
579:
580: Revision 1.112 2006/01/30 09:55:26 brouard
581: (Module): Back to gnuplot.exe instead of wgnuplot.exe
582:
583: Revision 1.111 2006/01/25 20:38:18 brouard
584: (Module): Lots of cleaning and bugs added (Gompertz)
585: (Module): Comments can be added in data file. Missing date values
586: can be a simple dot '.'.
587:
588: Revision 1.110 2006/01/25 00:51:50 brouard
589: (Module): Lots of cleaning and bugs added (Gompertz)
590:
591: Revision 1.109 2006/01/24 19:37:15 brouard
592: (Module): Comments (lines starting with a #) are allowed in data.
593:
594: Revision 1.108 2006/01/19 18:05:42 lievre
595: Gnuplot problem appeared...
596: To be fixed
597:
598: Revision 1.107 2006/01/19 16:20:37 brouard
599: Test existence of gnuplot in imach path
600:
601: Revision 1.106 2006/01/19 13:24:36 brouard
602: Some cleaning and links added in html output
603:
604: Revision 1.105 2006/01/05 20:23:19 lievre
605: *** empty log message ***
606:
607: Revision 1.104 2005/09/30 16:11:43 lievre
608: (Module): sump fixed, loop imx fixed, and simplifications.
609: (Module): If the status is missing at the last wave but we know
610: that the person is alive, then we can code his/her status as -2
611: (instead of missing=-1 in earlier versions) and his/her
612: contributions to the likelihood is 1 - Prob of dying from last
613: health status (= 1-p13= p11+p12 in the easiest case of somebody in
614: the healthy state at last known wave). Version is 0.98
615:
616: Revision 1.103 2005/09/30 15:54:49 lievre
617: (Module): sump fixed, loop imx fixed, and simplifications.
618:
619: Revision 1.102 2004/09/15 17:31:30 brouard
620: Add the possibility to read data file including tab characters.
621:
622: Revision 1.101 2004/09/15 10:38:38 brouard
623: Fix on curr_time
624:
625: Revision 1.100 2004/07/12 18:29:06 brouard
626: Add version for Mac OS X. Just define UNIX in Makefile
627:
628: Revision 1.99 2004/06/05 08:57:40 brouard
629: *** empty log message ***
630:
631: Revision 1.98 2004/05/16 15:05:56 brouard
632: New version 0.97 . First attempt to estimate force of mortality
633: directly from the data i.e. without the need of knowing the health
634: state at each age, but using a Gompertz model: log u =a + b*age .
635: This is the basic analysis of mortality and should be done before any
636: other analysis, in order to test if the mortality estimated from the
637: cross-longitudinal survey is different from the mortality estimated
638: from other sources like vital statistic data.
639:
640: The same imach parameter file can be used but the option for mle should be -3.
641:
1.133 brouard 642: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 643: former routines in order to include the new code within the former code.
644:
645: The output is very simple: only an estimate of the intercept and of
646: the slope with 95% confident intervals.
647:
648: Current limitations:
649: A) Even if you enter covariates, i.e. with the
650: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
651: B) There is no computation of Life Expectancy nor Life Table.
652:
653: Revision 1.97 2004/02/20 13:25:42 lievre
654: Version 0.96d. Population forecasting command line is (temporarily)
655: suppressed.
656:
657: Revision 1.96 2003/07/15 15:38:55 brouard
658: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
659: rewritten within the same printf. Workaround: many printfs.
660:
661: Revision 1.95 2003/07/08 07:54:34 brouard
662: * imach.c (Repository):
663: (Repository): Using imachwizard code to output a more meaningful covariance
664: matrix (cov(a12,c31) instead of numbers.
665:
666: Revision 1.94 2003/06/27 13:00:02 brouard
667: Just cleaning
668:
669: Revision 1.93 2003/06/25 16:33:55 brouard
670: (Module): On windows (cygwin) function asctime_r doesn't
671: exist so I changed back to asctime which exists.
672: (Module): Version 0.96b
673:
674: Revision 1.92 2003/06/25 16:30:45 brouard
675: (Module): On windows (cygwin) function asctime_r doesn't
676: exist so I changed back to asctime which exists.
677:
678: Revision 1.91 2003/06/25 15:30:29 brouard
679: * imach.c (Repository): Duplicated warning errors corrected.
680: (Repository): Elapsed time after each iteration is now output. It
681: helps to forecast when convergence will be reached. Elapsed time
682: is stamped in powell. We created a new html file for the graphs
683: concerning matrix of covariance. It has extension -cov.htm.
684:
685: Revision 1.90 2003/06/24 12:34:15 brouard
686: (Module): Some bugs corrected for windows. Also, when
687: mle=-1 a template is output in file "or"mypar.txt with the design
688: of the covariance matrix to be input.
689:
690: Revision 1.89 2003/06/24 12:30:52 brouard
691: (Module): Some bugs corrected for windows. Also, when
692: mle=-1 a template is output in file "or"mypar.txt with the design
693: of the covariance matrix to be input.
694:
695: Revision 1.88 2003/06/23 17:54:56 brouard
696: * 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.
697:
698: Revision 1.87 2003/06/18 12:26:01 brouard
699: Version 0.96
700:
701: Revision 1.86 2003/06/17 20:04:08 brouard
702: (Module): Change position of html and gnuplot routines and added
703: routine fileappend.
704:
705: Revision 1.85 2003/06/17 13:12:43 brouard
706: * imach.c (Repository): Check when date of death was earlier that
707: current date of interview. It may happen when the death was just
708: prior to the death. In this case, dh was negative and likelihood
709: was wrong (infinity). We still send an "Error" but patch by
710: assuming that the date of death was just one stepm after the
711: interview.
712: (Repository): Because some people have very long ID (first column)
713: we changed int to long in num[] and we added a new lvector for
714: memory allocation. But we also truncated to 8 characters (left
715: truncation)
716: (Repository): No more line truncation errors.
717:
718: Revision 1.84 2003/06/13 21:44:43 brouard
719: * imach.c (Repository): Replace "freqsummary" at a correct
720: place. It differs from routine "prevalence" which may be called
721: many times. Probs is memory consuming and must be used with
722: parcimony.
723: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
724:
725: Revision 1.83 2003/06/10 13:39:11 lievre
726: *** empty log message ***
727:
728: Revision 1.82 2003/06/05 15:57:20 brouard
729: Add log in imach.c and fullversion number is now printed.
730:
731: */
732: /*
733: Interpolated Markov Chain
734:
735: Short summary of the programme:
736:
1.227 brouard 737: This program computes Healthy Life Expectancies or State-specific
738: (if states aren't health statuses) Expectancies from
739: cross-longitudinal data. Cross-longitudinal data consist in:
740:
741: -1- a first survey ("cross") where individuals from different ages
742: are interviewed on their health status or degree of disability (in
743: the case of a health survey which is our main interest)
744:
745: -2- at least a second wave of interviews ("longitudinal") which
746: measure each change (if any) in individual health status. Health
747: expectancies are computed from the time spent in each health state
748: according to a model. More health states you consider, more time is
749: necessary to reach the Maximum Likelihood of the parameters involved
750: in the model. The simplest model is the multinomial logistic model
751: where pij is the probability to be observed in state j at the second
752: wave conditional to be observed in state i at the first
753: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
754: etc , where 'age' is age and 'sex' is a covariate. If you want to
755: have a more complex model than "constant and age", you should modify
756: the program where the markup *Covariates have to be included here
757: again* invites you to do it. More covariates you add, slower the
1.126 brouard 758: convergence.
759:
760: The advantage of this computer programme, compared to a simple
761: multinomial logistic model, is clear when the delay between waves is not
762: identical for each individual. Also, if a individual missed an
763: intermediate interview, the information is lost, but taken into
764: account using an interpolation or extrapolation.
765:
766: hPijx is the probability to be observed in state i at age x+h
767: conditional to the observed state i at age x. The delay 'h' can be
768: split into an exact number (nh*stepm) of unobserved intermediate
769: states. This elementary transition (by month, quarter,
770: semester or year) is modelled as a multinomial logistic. The hPx
771: matrix is simply the matrix product of nh*stepm elementary matrices
772: and the contribution of each individual to the likelihood is simply
773: hPijx.
774:
775: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 776: of the life expectancies. It also computes the period (stable) prevalence.
777:
778: Back prevalence and projections:
1.227 brouard 779:
780: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
781: double agemaxpar, double ftolpl, int *ncvyearp, double
782: dateprev1,double dateprev2, int firstpass, int lastpass, int
783: mobilavproj)
784:
785: Computes the back prevalence limit for any combination of
786: covariate values k at any age between ageminpar and agemaxpar and
787: returns it in **bprlim. In the loops,
788:
789: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
790: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
791:
792: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 793: Computes for any combination of covariates k and any age between bage and fage
794: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
795: oldm=oldms;savm=savms;
1.227 brouard 796:
797: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 798: Computes the transition matrix starting at age 'age' over
799: 'nhstepm*hstepm*stepm' months (i.e. until
800: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 801: nhstepm*hstepm matrices.
802:
803: Returns p3mat[i][j][h] after calling
804: p3mat[i][j][h]=matprod2(newm,
805: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
806: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
807: oldm);
1.226 brouard 808:
809: Important routines
810:
811: - func (or funcone), computes logit (pij) distinguishing
812: o fixed variables (single or product dummies or quantitative);
813: o varying variables by:
814: (1) wave (single, product dummies, quantitative),
815: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
816: % fixed dummy (treated) or quantitative (not done because time-consuming);
817: % varying dummy (not done) or quantitative (not done);
818: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
819: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
820: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
821: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
822: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 823:
1.226 brouard 824:
825:
1.133 brouard 826: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
827: Institut national d'études démographiques, Paris.
1.126 brouard 828: This software have been partly granted by Euro-REVES, a concerted action
829: from the European Union.
830: It is copyrighted identically to a GNU software product, ie programme and
831: software can be distributed freely for non commercial use. Latest version
832: can be accessed at http://euroreves.ined.fr/imach .
833:
834: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
835: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
836:
837: **********************************************************************/
838: /*
839: main
840: read parameterfile
841: read datafile
842: concatwav
843: freqsummary
844: if (mle >= 1)
845: mlikeli
846: print results files
847: if mle==1
848: computes hessian
849: read end of parameter file: agemin, agemax, bage, fage, estepm
850: begin-prev-date,...
851: open gnuplot file
852: open html file
1.145 brouard 853: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
854: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
855: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
856: freexexit2 possible for memory heap.
857:
858: h Pij x | pij_nom ficrestpij
859: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
860: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
861: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
862:
863: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
864: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
865: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
866: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
867: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
868:
1.126 brouard 869: forecasting if prevfcast==1 prevforecast call prevalence()
870: health expectancies
871: Variance-covariance of DFLE
872: prevalence()
873: movingaverage()
874: varevsij()
875: if popbased==1 varevsij(,popbased)
876: total life expectancies
877: Variance of period (stable) prevalence
878: end
879: */
880:
1.187 brouard 881: /* #define DEBUG */
882: /* #define DEBUGBRENT */
1.203 brouard 883: /* #define DEBUGLINMIN */
884: /* #define DEBUGHESS */
885: #define DEBUGHESSIJ
1.224 brouard 886: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 887: #define POWELL /* Instead of NLOPT */
1.224 brouard 888: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 889: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
890: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 891:
892: #include <math.h>
893: #include <stdio.h>
894: #include <stdlib.h>
895: #include <string.h>
1.226 brouard 896: #include <ctype.h>
1.159 brouard 897:
898: #ifdef _WIN32
899: #include <io.h>
1.172 brouard 900: #include <windows.h>
901: #include <tchar.h>
1.159 brouard 902: #else
1.126 brouard 903: #include <unistd.h>
1.159 brouard 904: #endif
1.126 brouard 905:
906: #include <limits.h>
907: #include <sys/types.h>
1.171 brouard 908:
909: #if defined(__GNUC__)
910: #include <sys/utsname.h> /* Doesn't work on Windows */
911: #endif
912:
1.126 brouard 913: #include <sys/stat.h>
914: #include <errno.h>
1.159 brouard 915: /* extern int errno; */
1.126 brouard 916:
1.157 brouard 917: /* #ifdef LINUX */
918: /* #include <time.h> */
919: /* #include "timeval.h" */
920: /* #else */
921: /* #include <sys/time.h> */
922: /* #endif */
923:
1.126 brouard 924: #include <time.h>
925:
1.136 brouard 926: #ifdef GSL
927: #include <gsl/gsl_errno.h>
928: #include <gsl/gsl_multimin.h>
929: #endif
930:
1.167 brouard 931:
1.162 brouard 932: #ifdef NLOPT
933: #include <nlopt.h>
934: typedef struct {
935: double (* function)(double [] );
936: } myfunc_data ;
937: #endif
938:
1.126 brouard 939: /* #include <libintl.h> */
940: /* #define _(String) gettext (String) */
941:
1.251 brouard 942: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 943:
944: #define GNUPLOTPROGRAM "gnuplot"
945: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
946: #define FILENAMELENGTH 132
947:
948: #define GLOCK_ERROR_NOPATH -1 /* empty path */
949: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
950:
1.144 brouard 951: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
952: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 953:
954: #define NINTERVMAX 8
1.144 brouard 955: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
956: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
957: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 958: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 959: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
960: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 961: #define MAXN 20000
1.144 brouard 962: #define YEARM 12. /**< Number of months per year */
1.218 brouard 963: /* #define AGESUP 130 */
964: #define AGESUP 150
965: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 966: #define AGEBASE 40
1.194 brouard 967: #define AGEOVERFLOW 1.e20
1.164 brouard 968: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 969: #ifdef _WIN32
970: #define DIRSEPARATOR '\\'
971: #define CHARSEPARATOR "\\"
972: #define ODIRSEPARATOR '/'
973: #else
1.126 brouard 974: #define DIRSEPARATOR '/'
975: #define CHARSEPARATOR "/"
976: #define ODIRSEPARATOR '\\'
977: #endif
978:
1.259 ! brouard 979: /* $Id: imach.c,v 1.258 2017/04/03 10:17:47 brouard Exp $ */
1.126 brouard 980: /* $State: Exp $ */
1.196 brouard 981: #include "version.h"
982: char version[]=__IMACH_VERSION__;
1.224 brouard 983: 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";
1.259 ! brouard 984: char fullversion[]="$Revision: 1.258 $ $Date: 2017/04/03 10:17:47 $";
1.126 brouard 985: char strstart[80];
986: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 987: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 988: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 989: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
990: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
991: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 992: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
993: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 994: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
995: int cptcovprodnoage=0; /**< Number of covariate products without age */
996: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 997: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
998: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 999: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1000: int nsd=0; /**< Total number of single dummy variables (output) */
1001: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1002: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1003: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1004: int ntveff=0; /**< ntveff number of effective time varying variables */
1005: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1006: int cptcov=0; /* Working variable */
1.218 brouard 1007: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1008: int npar=NPARMAX;
1009: int nlstate=2; /* Number of live states */
1010: int ndeath=1; /* Number of dead states */
1.130 brouard 1011: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1012: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1013: int popbased=0;
1014:
1015: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1016: int maxwav=0; /* Maxim number of waves */
1017: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1018: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1019: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1020: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1021: int mle=1, weightopt=0;
1.126 brouard 1022: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1023: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1024: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1025: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1026: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1027: int selected(int kvar); /* Is covariate kvar selected for printing results */
1028:
1.130 brouard 1029: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1030: double **matprod2(); /* test */
1.126 brouard 1031: double **oldm, **newm, **savm; /* Working pointers to matrices */
1032: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1033: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1034:
1.136 brouard 1035: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1036: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1037: FILE *ficlog, *ficrespow;
1.130 brouard 1038: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1039: double fretone; /* Only one call to likelihood */
1.130 brouard 1040: long ipmx=0; /* Number of contributions */
1.126 brouard 1041: double sw; /* Sum of weights */
1042: char filerespow[FILENAMELENGTH];
1043: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1044: FILE *ficresilk;
1045: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1046: FILE *ficresprobmorprev;
1047: FILE *fichtm, *fichtmcov; /* Html File */
1048: FILE *ficreseij;
1049: char filerese[FILENAMELENGTH];
1050: FILE *ficresstdeij;
1051: char fileresstde[FILENAMELENGTH];
1052: FILE *ficrescveij;
1053: char filerescve[FILENAMELENGTH];
1054: FILE *ficresvij;
1055: char fileresv[FILENAMELENGTH];
1056: FILE *ficresvpl;
1057: char fileresvpl[FILENAMELENGTH];
1058: char title[MAXLINE];
1.234 brouard 1059: char model[MAXLINE]; /**< The model line */
1.217 brouard 1060: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1061: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1062: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1063: char command[FILENAMELENGTH];
1064: int outcmd=0;
1065:
1.217 brouard 1066: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1067: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1068: char filelog[FILENAMELENGTH]; /* Log file */
1069: char filerest[FILENAMELENGTH];
1070: char fileregp[FILENAMELENGTH];
1071: char popfile[FILENAMELENGTH];
1072:
1073: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1074:
1.157 brouard 1075: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1076: /* struct timezone tzp; */
1077: /* extern int gettimeofday(); */
1078: struct tm tml, *gmtime(), *localtime();
1079:
1080: extern time_t time();
1081:
1082: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1083: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1084: struct tm tm;
1085:
1.126 brouard 1086: char strcurr[80], strfor[80];
1087:
1088: char *endptr;
1089: long lval;
1090: double dval;
1091:
1092: #define NR_END 1
1093: #define FREE_ARG char*
1094: #define FTOL 1.0e-10
1095:
1096: #define NRANSI
1.240 brouard 1097: #define ITMAX 200
1098: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1099:
1100: #define TOL 2.0e-4
1101:
1102: #define CGOLD 0.3819660
1103: #define ZEPS 1.0e-10
1104: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1105:
1106: #define GOLD 1.618034
1107: #define GLIMIT 100.0
1108: #define TINY 1.0e-20
1109:
1110: static double maxarg1,maxarg2;
1111: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1112: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1113:
1114: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1115: #define rint(a) floor(a+0.5)
1.166 brouard 1116: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1117: #define mytinydouble 1.0e-16
1.166 brouard 1118: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1119: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1120: /* static double dsqrarg; */
1121: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1122: static double sqrarg;
1123: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1124: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1125: int agegomp= AGEGOMP;
1126:
1127: int imx;
1128: int stepm=1;
1129: /* Stepm, step in month: minimum step interpolation*/
1130:
1131: int estepm;
1132: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1133:
1134: int m,nb;
1135: long *num;
1.197 brouard 1136: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1137: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1138: covariate for which somebody answered excluding
1139: undefined. Usually 2: 0 and 1. */
1140: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1141: covariate for which somebody answered including
1142: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1143: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1144: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1145: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1146: double *ageexmed,*agecens;
1147: double dateintmean=0;
1148:
1149: double *weight;
1150: int **s; /* Status */
1.141 brouard 1151: double *agedc;
1.145 brouard 1152: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1153: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1154: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1155: double **coqvar; /* Fixed quantitative covariate iqv */
1156: double ***cotvar; /* Time varying covariate itv */
1157: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1158: double idx;
1159: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1160: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1161: /*k 1 2 3 4 5 6 7 8 9 */
1162: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1163: /* Tndvar[k] 1 2 3 4 5 */
1164: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1165: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1166: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1167: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1168: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1169: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1170: /* Tprod[i]=k 4 7 */
1171: /* Tage[i]=k 5 8 */
1172: /* */
1173: /* Type */
1174: /* V 1 2 3 4 5 */
1175: /* F F V V V */
1176: /* D Q D D Q */
1177: /* */
1178: int *TvarsD;
1179: int *TvarsDind;
1180: int *TvarsQ;
1181: int *TvarsQind;
1182:
1.235 brouard 1183: #define MAXRESULTLINES 10
1184: int nresult=0;
1.258 brouard 1185: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1186: int TKresult[MAXRESULTLINES];
1.237 brouard 1187: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1188: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1189: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1190: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1191: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1192: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1193:
1.234 brouard 1194: /* 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 1195: 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 */
1196: 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 */
1197: 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 */
1198: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1199: 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 */
1200: 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 1201: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1202: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1203: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1204: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1205: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1206: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1207: 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 */
1208: 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 */
1209:
1.230 brouard 1210: int *Tvarsel; /**< Selected covariates for output */
1211: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1212: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1213: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1214: 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 1215: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1216: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1217: int *Tage;
1.227 brouard 1218: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1219: 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 1220: 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*/
1221: 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 1222: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1223: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1224: int **Tvard;
1225: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1226: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1227: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1228: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1229: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1230: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1231: double *lsurv, *lpop, *tpop;
1232:
1.231 brouard 1233: #define FD 1; /* Fixed dummy covariate */
1234: #define FQ 2; /* Fixed quantitative covariate */
1235: #define FP 3; /* Fixed product covariate */
1236: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1237: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1238: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1239: #define VD 10; /* Varying dummy covariate */
1240: #define VQ 11; /* Varying quantitative covariate */
1241: #define VP 12; /* Varying product covariate */
1242: #define VPDD 13; /* Varying product dummy*dummy covariate */
1243: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1244: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1245: #define APFD 16; /* Age product * fixed dummy covariate */
1246: #define APFQ 17; /* Age product * fixed quantitative covariate */
1247: #define APVD 18; /* Age product * varying dummy covariate */
1248: #define APVQ 19; /* Age product * varying quantitative covariate */
1249:
1250: #define FTYPE 1; /* Fixed covariate */
1251: #define VTYPE 2; /* Varying covariate (loop in wave) */
1252: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1253:
1254: struct kmodel{
1255: int maintype; /* main type */
1256: int subtype; /* subtype */
1257: };
1258: struct kmodel modell[NCOVMAX];
1259:
1.143 brouard 1260: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1261: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1262:
1263: /**************** split *************************/
1264: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1265: {
1266: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1267: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1268: */
1269: char *ss; /* pointer */
1.186 brouard 1270: int l1=0, l2=0; /* length counters */
1.126 brouard 1271:
1272: l1 = strlen(path ); /* length of path */
1273: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1274: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1275: if ( ss == NULL ) { /* no directory, so determine current directory */
1276: strcpy( name, path ); /* we got the fullname name because no directory */
1277: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1278: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1279: /* get current working directory */
1280: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1281: #ifdef WIN32
1282: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1283: #else
1284: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1285: #endif
1.126 brouard 1286: return( GLOCK_ERROR_GETCWD );
1287: }
1288: /* got dirc from getcwd*/
1289: printf(" DIRC = %s \n",dirc);
1.205 brouard 1290: } else { /* strip directory from path */
1.126 brouard 1291: ss++; /* after this, the filename */
1292: l2 = strlen( ss ); /* length of filename */
1293: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1294: strcpy( name, ss ); /* save file name */
1295: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1296: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1297: printf(" DIRC2 = %s \n",dirc);
1298: }
1299: /* We add a separator at the end of dirc if not exists */
1300: l1 = strlen( dirc ); /* length of directory */
1301: if( dirc[l1-1] != DIRSEPARATOR ){
1302: dirc[l1] = DIRSEPARATOR;
1303: dirc[l1+1] = 0;
1304: printf(" DIRC3 = %s \n",dirc);
1305: }
1306: ss = strrchr( name, '.' ); /* find last / */
1307: if (ss >0){
1308: ss++;
1309: strcpy(ext,ss); /* save extension */
1310: l1= strlen( name);
1311: l2= strlen(ss)+1;
1312: strncpy( finame, name, l1-l2);
1313: finame[l1-l2]= 0;
1314: }
1315:
1316: return( 0 ); /* we're done */
1317: }
1318:
1319:
1320: /******************************************/
1321:
1322: void replace_back_to_slash(char *s, char*t)
1323: {
1324: int i;
1325: int lg=0;
1326: i=0;
1327: lg=strlen(t);
1328: for(i=0; i<= lg; i++) {
1329: (s[i] = t[i]);
1330: if (t[i]== '\\') s[i]='/';
1331: }
1332: }
1333:
1.132 brouard 1334: char *trimbb(char *out, char *in)
1.137 brouard 1335: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1336: char *s;
1337: s=out;
1338: while (*in != '\0'){
1.137 brouard 1339: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1340: in++;
1341: }
1342: *out++ = *in++;
1343: }
1344: *out='\0';
1345: return s;
1346: }
1347:
1.187 brouard 1348: /* char *substrchaine(char *out, char *in, char *chain) */
1349: /* { */
1350: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1351: /* char *s, *t; */
1352: /* t=in;s=out; */
1353: /* while ((*in != *chain) && (*in != '\0')){ */
1354: /* *out++ = *in++; */
1355: /* } */
1356:
1357: /* /\* *in matches *chain *\/ */
1358: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1359: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1360: /* } */
1361: /* in--; chain--; */
1362: /* while ( (*in != '\0')){ */
1363: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1364: /* *out++ = *in++; */
1365: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1366: /* } */
1367: /* *out='\0'; */
1368: /* out=s; */
1369: /* return out; */
1370: /* } */
1371: char *substrchaine(char *out, char *in, char *chain)
1372: {
1373: /* Substract chain 'chain' from 'in', return and output 'out' */
1374: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1375:
1376: char *strloc;
1377:
1378: strcpy (out, in);
1379: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1380: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1381: if(strloc != NULL){
1382: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1383: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1384: /* strcpy (strloc, strloc +strlen(chain));*/
1385: }
1386: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1387: return out;
1388: }
1389:
1390:
1.145 brouard 1391: char *cutl(char *blocc, char *alocc, char *in, char occ)
1392: {
1.187 brouard 1393: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1394: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1395: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1396: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1397: */
1.160 brouard 1398: char *s, *t;
1.145 brouard 1399: t=in;s=in;
1400: while ((*in != occ) && (*in != '\0')){
1401: *alocc++ = *in++;
1402: }
1403: if( *in == occ){
1404: *(alocc)='\0';
1405: s=++in;
1406: }
1407:
1408: if (s == t) {/* occ not found */
1409: *(alocc-(in-s))='\0';
1410: in=s;
1411: }
1412: while ( *in != '\0'){
1413: *blocc++ = *in++;
1414: }
1415:
1416: *blocc='\0';
1417: return t;
1418: }
1.137 brouard 1419: char *cutv(char *blocc, char *alocc, char *in, char occ)
1420: {
1.187 brouard 1421: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1422: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1423: gives blocc="abcdef2ghi" and alocc="j".
1424: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1425: */
1426: char *s, *t;
1427: t=in;s=in;
1428: while (*in != '\0'){
1429: while( *in == occ){
1430: *blocc++ = *in++;
1431: s=in;
1432: }
1433: *blocc++ = *in++;
1434: }
1435: if (s == t) /* occ not found */
1436: *(blocc-(in-s))='\0';
1437: else
1438: *(blocc-(in-s)-1)='\0';
1439: in=s;
1440: while ( *in != '\0'){
1441: *alocc++ = *in++;
1442: }
1443:
1444: *alocc='\0';
1445: return s;
1446: }
1447:
1.126 brouard 1448: int nbocc(char *s, char occ)
1449: {
1450: int i,j=0;
1451: int lg=20;
1452: i=0;
1453: lg=strlen(s);
1454: for(i=0; i<= lg; i++) {
1.234 brouard 1455: if (s[i] == occ ) j++;
1.126 brouard 1456: }
1457: return j;
1458: }
1459:
1.137 brouard 1460: /* void cutv(char *u,char *v, char*t, char occ) */
1461: /* { */
1462: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1463: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1464: /* gives u="abcdef2ghi" and v="j" *\/ */
1465: /* int i,lg,j,p=0; */
1466: /* i=0; */
1467: /* lg=strlen(t); */
1468: /* for(j=0; j<=lg-1; j++) { */
1469: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1470: /* } */
1.126 brouard 1471:
1.137 brouard 1472: /* for(j=0; j<p; j++) { */
1473: /* (u[j] = t[j]); */
1474: /* } */
1475: /* u[p]='\0'; */
1.126 brouard 1476:
1.137 brouard 1477: /* for(j=0; j<= lg; j++) { */
1478: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1479: /* } */
1480: /* } */
1.126 brouard 1481:
1.160 brouard 1482: #ifdef _WIN32
1483: char * strsep(char **pp, const char *delim)
1484: {
1485: char *p, *q;
1486:
1487: if ((p = *pp) == NULL)
1488: return 0;
1489: if ((q = strpbrk (p, delim)) != NULL)
1490: {
1491: *pp = q + 1;
1492: *q = '\0';
1493: }
1494: else
1495: *pp = 0;
1496: return p;
1497: }
1498: #endif
1499:
1.126 brouard 1500: /********************** nrerror ********************/
1501:
1502: void nrerror(char error_text[])
1503: {
1504: fprintf(stderr,"ERREUR ...\n");
1505: fprintf(stderr,"%s\n",error_text);
1506: exit(EXIT_FAILURE);
1507: }
1508: /*********************** vector *******************/
1509: double *vector(int nl, int nh)
1510: {
1511: double *v;
1512: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1513: if (!v) nrerror("allocation failure in vector");
1514: return v-nl+NR_END;
1515: }
1516:
1517: /************************ free vector ******************/
1518: void free_vector(double*v, int nl, int nh)
1519: {
1520: free((FREE_ARG)(v+nl-NR_END));
1521: }
1522:
1523: /************************ivector *******************************/
1524: int *ivector(long nl,long nh)
1525: {
1526: int *v;
1527: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1528: if (!v) nrerror("allocation failure in ivector");
1529: return v-nl+NR_END;
1530: }
1531:
1532: /******************free ivector **************************/
1533: void free_ivector(int *v, long nl, long nh)
1534: {
1535: free((FREE_ARG)(v+nl-NR_END));
1536: }
1537:
1538: /************************lvector *******************************/
1539: long *lvector(long nl,long nh)
1540: {
1541: long *v;
1542: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1543: if (!v) nrerror("allocation failure in ivector");
1544: return v-nl+NR_END;
1545: }
1546:
1547: /******************free lvector **************************/
1548: void free_lvector(long *v, long nl, long nh)
1549: {
1550: free((FREE_ARG)(v+nl-NR_END));
1551: }
1552:
1553: /******************* imatrix *******************************/
1554: int **imatrix(long nrl, long nrh, long ncl, long nch)
1555: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1556: {
1557: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1558: int **m;
1559:
1560: /* allocate pointers to rows */
1561: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1562: if (!m) nrerror("allocation failure 1 in matrix()");
1563: m += NR_END;
1564: m -= nrl;
1565:
1566:
1567: /* allocate rows and set pointers to them */
1568: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1569: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1570: m[nrl] += NR_END;
1571: m[nrl] -= ncl;
1572:
1573: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1574:
1575: /* return pointer to array of pointers to rows */
1576: return m;
1577: }
1578:
1579: /****************** free_imatrix *************************/
1580: void free_imatrix(m,nrl,nrh,ncl,nch)
1581: int **m;
1582: long nch,ncl,nrh,nrl;
1583: /* free an int matrix allocated by imatrix() */
1584: {
1585: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1586: free((FREE_ARG) (m+nrl-NR_END));
1587: }
1588:
1589: /******************* matrix *******************************/
1590: double **matrix(long nrl, long nrh, long ncl, long nch)
1591: {
1592: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1593: double **m;
1594:
1595: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1596: if (!m) nrerror("allocation failure 1 in matrix()");
1597: m += NR_END;
1598: m -= nrl;
1599:
1600: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1601: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1602: m[nrl] += NR_END;
1603: m[nrl] -= ncl;
1604:
1605: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1606: return m;
1.145 brouard 1607: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1608: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1609: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1610: */
1611: }
1612:
1613: /*************************free matrix ************************/
1614: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1615: {
1616: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1617: free((FREE_ARG)(m+nrl-NR_END));
1618: }
1619:
1620: /******************* ma3x *******************************/
1621: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1622: {
1623: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1624: double ***m;
1625:
1626: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1627: if (!m) nrerror("allocation failure 1 in matrix()");
1628: m += NR_END;
1629: m -= nrl;
1630:
1631: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1632: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1633: m[nrl] += NR_END;
1634: m[nrl] -= ncl;
1635:
1636: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1637:
1638: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1639: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1640: m[nrl][ncl] += NR_END;
1641: m[nrl][ncl] -= nll;
1642: for (j=ncl+1; j<=nch; j++)
1643: m[nrl][j]=m[nrl][j-1]+nlay;
1644:
1645: for (i=nrl+1; i<=nrh; i++) {
1646: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1647: for (j=ncl+1; j<=nch; j++)
1648: m[i][j]=m[i][j-1]+nlay;
1649: }
1650: return m;
1651: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1652: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1653: */
1654: }
1655:
1656: /*************************free ma3x ************************/
1657: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1658: {
1659: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1660: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1661: free((FREE_ARG)(m+nrl-NR_END));
1662: }
1663:
1664: /*************** function subdirf ***********/
1665: char *subdirf(char fileres[])
1666: {
1667: /* Caution optionfilefiname is hidden */
1668: strcpy(tmpout,optionfilefiname);
1669: strcat(tmpout,"/"); /* Add to the right */
1670: strcat(tmpout,fileres);
1671: return tmpout;
1672: }
1673:
1674: /*************** function subdirf2 ***********/
1675: char *subdirf2(char fileres[], char *preop)
1676: {
1677:
1678: /* Caution optionfilefiname is hidden */
1679: strcpy(tmpout,optionfilefiname);
1680: strcat(tmpout,"/");
1681: strcat(tmpout,preop);
1682: strcat(tmpout,fileres);
1683: return tmpout;
1684: }
1685:
1686: /*************** function subdirf3 ***********/
1687: char *subdirf3(char fileres[], char *preop, char *preop2)
1688: {
1689:
1690: /* Caution optionfilefiname is hidden */
1691: strcpy(tmpout,optionfilefiname);
1692: strcat(tmpout,"/");
1693: strcat(tmpout,preop);
1694: strcat(tmpout,preop2);
1695: strcat(tmpout,fileres);
1696: return tmpout;
1697: }
1.213 brouard 1698:
1699: /*************** function subdirfext ***********/
1700: char *subdirfext(char fileres[], char *preop, char *postop)
1701: {
1702:
1703: strcpy(tmpout,preop);
1704: strcat(tmpout,fileres);
1705: strcat(tmpout,postop);
1706: return tmpout;
1707: }
1.126 brouard 1708:
1.213 brouard 1709: /*************** function subdirfext3 ***********/
1710: char *subdirfext3(char fileres[], char *preop, char *postop)
1711: {
1712:
1713: /* Caution optionfilefiname is hidden */
1714: strcpy(tmpout,optionfilefiname);
1715: strcat(tmpout,"/");
1716: strcat(tmpout,preop);
1717: strcat(tmpout,fileres);
1718: strcat(tmpout,postop);
1719: return tmpout;
1720: }
1721:
1.162 brouard 1722: char *asc_diff_time(long time_sec, char ascdiff[])
1723: {
1724: long sec_left, days, hours, minutes;
1725: days = (time_sec) / (60*60*24);
1726: sec_left = (time_sec) % (60*60*24);
1727: hours = (sec_left) / (60*60) ;
1728: sec_left = (sec_left) %(60*60);
1729: minutes = (sec_left) /60;
1730: sec_left = (sec_left) % (60);
1731: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1732: return ascdiff;
1733: }
1734:
1.126 brouard 1735: /***************** f1dim *************************/
1736: extern int ncom;
1737: extern double *pcom,*xicom;
1738: extern double (*nrfunc)(double []);
1739:
1740: double f1dim(double x)
1741: {
1742: int j;
1743: double f;
1744: double *xt;
1745:
1746: xt=vector(1,ncom);
1747: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1748: f=(*nrfunc)(xt);
1749: free_vector(xt,1,ncom);
1750: return f;
1751: }
1752:
1753: /*****************brent *************************/
1754: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1755: {
1756: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1757: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1758: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1759: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1760: * returned function value.
1761: */
1.126 brouard 1762: int iter;
1763: double a,b,d,etemp;
1.159 brouard 1764: double fu=0,fv,fw,fx;
1.164 brouard 1765: double ftemp=0.;
1.126 brouard 1766: double p,q,r,tol1,tol2,u,v,w,x,xm;
1767: double e=0.0;
1768:
1769: a=(ax < cx ? ax : cx);
1770: b=(ax > cx ? ax : cx);
1771: x=w=v=bx;
1772: fw=fv=fx=(*f)(x);
1773: for (iter=1;iter<=ITMAX;iter++) {
1774: xm=0.5*(a+b);
1775: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1776: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1777: printf(".");fflush(stdout);
1778: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1779: #ifdef DEBUGBRENT
1.126 brouard 1780: 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);
1781: 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);
1782: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1783: #endif
1784: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1785: *xmin=x;
1786: return fx;
1787: }
1788: ftemp=fu;
1789: if (fabs(e) > tol1) {
1790: r=(x-w)*(fx-fv);
1791: q=(x-v)*(fx-fw);
1792: p=(x-v)*q-(x-w)*r;
1793: q=2.0*(q-r);
1794: if (q > 0.0) p = -p;
1795: q=fabs(q);
1796: etemp=e;
1797: e=d;
1798: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1799: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1800: else {
1.224 brouard 1801: d=p/q;
1802: u=x+d;
1803: if (u-a < tol2 || b-u < tol2)
1804: d=SIGN(tol1,xm-x);
1.126 brouard 1805: }
1806: } else {
1807: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1808: }
1809: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1810: fu=(*f)(u);
1811: if (fu <= fx) {
1812: if (u >= x) a=x; else b=x;
1813: SHFT(v,w,x,u)
1.183 brouard 1814: SHFT(fv,fw,fx,fu)
1815: } else {
1816: if (u < x) a=u; else b=u;
1817: if (fu <= fw || w == x) {
1.224 brouard 1818: v=w;
1819: w=u;
1820: fv=fw;
1821: fw=fu;
1.183 brouard 1822: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1823: v=u;
1824: fv=fu;
1.183 brouard 1825: }
1826: }
1.126 brouard 1827: }
1828: nrerror("Too many iterations in brent");
1829: *xmin=x;
1830: return fx;
1831: }
1832:
1833: /****************** mnbrak ***********************/
1834:
1835: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1836: double (*func)(double))
1.183 brouard 1837: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1838: the downhill direction (defined by the function as evaluated at the initial points) and returns
1839: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1840: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1841: */
1.126 brouard 1842: double ulim,u,r,q, dum;
1843: double fu;
1.187 brouard 1844:
1845: double scale=10.;
1846: int iterscale=0;
1847:
1848: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1849: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1850:
1851:
1852: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1853: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1854: /* *bx = *ax - (*ax - *bx)/scale; */
1855: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1856: /* } */
1857:
1.126 brouard 1858: if (*fb > *fa) {
1859: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1860: SHFT(dum,*fb,*fa,dum)
1861: }
1.126 brouard 1862: *cx=(*bx)+GOLD*(*bx-*ax);
1863: *fc=(*func)(*cx);
1.183 brouard 1864: #ifdef DEBUG
1.224 brouard 1865: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1866: 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 1867: #endif
1.224 brouard 1868: 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 1869: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1870: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1871: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1872: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1873: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1874: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1875: fu=(*func)(u);
1.163 brouard 1876: #ifdef DEBUG
1877: /* f(x)=A(x-u)**2+f(u) */
1878: double A, fparabu;
1879: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1880: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1881: 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);
1882: 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 1883: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1884: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1885: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1886: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1887: #endif
1.184 brouard 1888: #ifdef MNBRAKORIGINAL
1.183 brouard 1889: #else
1.191 brouard 1890: /* if (fu > *fc) { */
1891: /* #ifdef DEBUG */
1892: /* printf("mnbrak4 fu > fc \n"); */
1893: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1894: /* #endif */
1895: /* /\* 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 *\\/ *\/ */
1896: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1897: /* dum=u; /\* Shifting c and u *\/ */
1898: /* u = *cx; */
1899: /* *cx = dum; */
1900: /* dum = fu; */
1901: /* fu = *fc; */
1902: /* *fc =dum; */
1903: /* } else { /\* end *\/ */
1904: /* #ifdef DEBUG */
1905: /* printf("mnbrak3 fu < fc \n"); */
1906: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1907: /* #endif */
1908: /* dum=u; /\* Shifting c and u *\/ */
1909: /* u = *cx; */
1910: /* *cx = dum; */
1911: /* dum = fu; */
1912: /* fu = *fc; */
1913: /* *fc =dum; */
1914: /* } */
1.224 brouard 1915: #ifdef DEBUGMNBRAK
1916: double A, fparabu;
1917: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1918: fparabu= *fa - A*(*ax-u)*(*ax-u);
1919: 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);
1920: 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 1921: #endif
1.191 brouard 1922: dum=u; /* Shifting c and u */
1923: u = *cx;
1924: *cx = dum;
1925: dum = fu;
1926: fu = *fc;
1927: *fc =dum;
1.183 brouard 1928: #endif
1.162 brouard 1929: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1930: #ifdef DEBUG
1.224 brouard 1931: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1932: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1933: #endif
1.126 brouard 1934: fu=(*func)(u);
1935: if (fu < *fc) {
1.183 brouard 1936: #ifdef DEBUG
1.224 brouard 1937: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1938: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1939: #endif
1940: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1941: SHFT(*fb,*fc,fu,(*func)(u))
1942: #ifdef DEBUG
1943: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1944: #endif
1945: }
1.162 brouard 1946: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1947: #ifdef DEBUG
1.224 brouard 1948: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1949: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1950: #endif
1.126 brouard 1951: u=ulim;
1952: fu=(*func)(u);
1.183 brouard 1953: } else { /* u could be left to b (if r > q parabola has a maximum) */
1954: #ifdef DEBUG
1.224 brouard 1955: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1956: 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 1957: #endif
1.126 brouard 1958: u=(*cx)+GOLD*(*cx-*bx);
1959: fu=(*func)(u);
1.224 brouard 1960: #ifdef DEBUG
1961: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1962: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1963: #endif
1.183 brouard 1964: } /* end tests */
1.126 brouard 1965: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1966: SHFT(*fa,*fb,*fc,fu)
1967: #ifdef DEBUG
1.224 brouard 1968: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1969: 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 1970: #endif
1971: } /* 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 1972: }
1973:
1974: /*************** linmin ************************/
1.162 brouard 1975: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1976: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1977: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1978: the value of func at the returned location p . This is actually all accomplished by calling the
1979: routines mnbrak and brent .*/
1.126 brouard 1980: int ncom;
1981: double *pcom,*xicom;
1982: double (*nrfunc)(double []);
1983:
1.224 brouard 1984: #ifdef LINMINORIGINAL
1.126 brouard 1985: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1986: #else
1987: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1988: #endif
1.126 brouard 1989: {
1990: double brent(double ax, double bx, double cx,
1991: double (*f)(double), double tol, double *xmin);
1992: double f1dim(double x);
1993: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1994: double *fc, double (*func)(double));
1995: int j;
1996: double xx,xmin,bx,ax;
1997: double fx,fb,fa;
1.187 brouard 1998:
1.203 brouard 1999: #ifdef LINMINORIGINAL
2000: #else
2001: double scale=10., axs, xxs; /* Scale added for infinity */
2002: #endif
2003:
1.126 brouard 2004: ncom=n;
2005: pcom=vector(1,n);
2006: xicom=vector(1,n);
2007: nrfunc=func;
2008: for (j=1;j<=n;j++) {
2009: pcom[j]=p[j];
1.202 brouard 2010: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2011: }
1.187 brouard 2012:
1.203 brouard 2013: #ifdef LINMINORIGINAL
2014: xx=1.;
2015: #else
2016: axs=0.0;
2017: xxs=1.;
2018: do{
2019: xx= xxs;
2020: #endif
1.187 brouard 2021: ax=0.;
2022: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2023: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2024: /* 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)) */
2025: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2026: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2027: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2028: /* 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 2029: #ifdef LINMINORIGINAL
2030: #else
2031: if (fx != fx){
1.224 brouard 2032: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2033: printf("|");
2034: fprintf(ficlog,"|");
1.203 brouard 2035: #ifdef DEBUGLINMIN
1.224 brouard 2036: 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 2037: #endif
2038: }
1.224 brouard 2039: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2040: #endif
2041:
1.191 brouard 2042: #ifdef DEBUGLINMIN
2043: 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 2044: 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 2045: #endif
1.224 brouard 2046: #ifdef LINMINORIGINAL
2047: #else
2048: if(fb == fx){ /* Flat function in the direction */
2049: xmin=xx;
2050: *flat=1;
2051: }else{
2052: *flat=0;
2053: #endif
2054: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2055: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2056: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2057: /* fmin = f(p[j] + xmin * xi[j]) */
2058: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2059: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2060: #ifdef DEBUG
1.224 brouard 2061: 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);
2062: 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);
2063: #endif
2064: #ifdef LINMINORIGINAL
2065: #else
2066: }
1.126 brouard 2067: #endif
1.191 brouard 2068: #ifdef DEBUGLINMIN
2069: printf("linmin end ");
1.202 brouard 2070: fprintf(ficlog,"linmin end ");
1.191 brouard 2071: #endif
1.126 brouard 2072: for (j=1;j<=n;j++) {
1.203 brouard 2073: #ifdef LINMINORIGINAL
2074: xi[j] *= xmin;
2075: #else
2076: #ifdef DEBUGLINMIN
2077: if(xxs <1.0)
2078: printf(" before xi[%d]=%12.8f", j,xi[j]);
2079: #endif
2080: 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) */
2081: #ifdef DEBUGLINMIN
2082: if(xxs <1.0)
2083: 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 );
2084: #endif
2085: #endif
1.187 brouard 2086: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2087: }
1.191 brouard 2088: #ifdef DEBUGLINMIN
1.203 brouard 2089: printf("\n");
1.191 brouard 2090: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2091: 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 2092: for (j=1;j<=n;j++) {
1.202 brouard 2093: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2094: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2095: if(j % ncovmodel == 0){
1.191 brouard 2096: printf("\n");
1.202 brouard 2097: fprintf(ficlog,"\n");
2098: }
1.191 brouard 2099: }
1.203 brouard 2100: #else
1.191 brouard 2101: #endif
1.126 brouard 2102: free_vector(xicom,1,n);
2103: free_vector(pcom,1,n);
2104: }
2105:
2106:
2107: /*************** powell ************************/
1.162 brouard 2108: /*
2109: Minimization of a function func of n variables. Input consists of an initial starting point
2110: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2111: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2112: such that failure to decrease by more than this amount on one iteration signals doneness. On
2113: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2114: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2115: */
1.224 brouard 2116: #ifdef LINMINORIGINAL
2117: #else
2118: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2119: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2120: #endif
1.126 brouard 2121: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2122: double (*func)(double []))
2123: {
1.224 brouard 2124: #ifdef LINMINORIGINAL
2125: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2126: double (*func)(double []));
1.224 brouard 2127: #else
1.241 brouard 2128: void linmin(double p[], double xi[], int n, double *fret,
2129: double (*func)(double []),int *flat);
1.224 brouard 2130: #endif
1.239 brouard 2131: int i,ibig,j,jk,k;
1.126 brouard 2132: double del,t,*pt,*ptt,*xit;
1.181 brouard 2133: double directest;
1.126 brouard 2134: double fp,fptt;
2135: double *xits;
2136: int niterf, itmp;
1.224 brouard 2137: #ifdef LINMINORIGINAL
2138: #else
2139:
2140: flatdir=ivector(1,n);
2141: for (j=1;j<=n;j++) flatdir[j]=0;
2142: #endif
1.126 brouard 2143:
2144: pt=vector(1,n);
2145: ptt=vector(1,n);
2146: xit=vector(1,n);
2147: xits=vector(1,n);
2148: *fret=(*func)(p);
2149: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2150: rcurr_time = time(NULL);
1.126 brouard 2151: for (*iter=1;;++(*iter)) {
1.187 brouard 2152: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2153: ibig=0;
2154: del=0.0;
1.157 brouard 2155: rlast_time=rcurr_time;
2156: /* (void) gettimeofday(&curr_time,&tzp); */
2157: rcurr_time = time(NULL);
2158: curr_time = *localtime(&rcurr_time);
2159: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2160: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2161: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2162: for (i=1;i<=n;i++) {
1.126 brouard 2163: fprintf(ficrespow," %.12lf", p[i]);
2164: }
1.239 brouard 2165: fprintf(ficrespow,"\n");fflush(ficrespow);
2166: printf("\n#model= 1 + age ");
2167: fprintf(ficlog,"\n#model= 1 + age ");
2168: if(nagesqr==1){
1.241 brouard 2169: printf(" + age*age ");
2170: fprintf(ficlog," + age*age ");
1.239 brouard 2171: }
2172: for(j=1;j <=ncovmodel-2;j++){
2173: if(Typevar[j]==0) {
2174: printf(" + V%d ",Tvar[j]);
2175: fprintf(ficlog," + V%d ",Tvar[j]);
2176: }else if(Typevar[j]==1) {
2177: printf(" + V%d*age ",Tvar[j]);
2178: fprintf(ficlog," + V%d*age ",Tvar[j]);
2179: }else if(Typevar[j]==2) {
2180: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2181: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2182: }
2183: }
1.126 brouard 2184: printf("\n");
1.239 brouard 2185: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2186: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2187: fprintf(ficlog,"\n");
1.239 brouard 2188: for(i=1,jk=1; i <=nlstate; i++){
2189: for(k=1; k <=(nlstate+ndeath); k++){
2190: if (k != i) {
2191: printf("%d%d ",i,k);
2192: fprintf(ficlog,"%d%d ",i,k);
2193: for(j=1; j <=ncovmodel; j++){
2194: printf("%12.7f ",p[jk]);
2195: fprintf(ficlog,"%12.7f ",p[jk]);
2196: jk++;
2197: }
2198: printf("\n");
2199: fprintf(ficlog,"\n");
2200: }
2201: }
2202: }
1.241 brouard 2203: if(*iter <=3 && *iter >1){
1.157 brouard 2204: tml = *localtime(&rcurr_time);
2205: strcpy(strcurr,asctime(&tml));
2206: rforecast_time=rcurr_time;
1.126 brouard 2207: itmp = strlen(strcurr);
2208: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2209: strcurr[itmp-1]='\0';
1.162 brouard 2210: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2211: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2212: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2213: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2214: forecast_time = *localtime(&rforecast_time);
2215: strcpy(strfor,asctime(&forecast_time));
2216: itmp = strlen(strfor);
2217: if(strfor[itmp-1]=='\n')
2218: strfor[itmp-1]='\0';
2219: 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);
2220: 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 2221: }
2222: }
1.187 brouard 2223: for (i=1;i<=n;i++) { /* For each direction i */
2224: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2225: fptt=(*fret);
2226: #ifdef DEBUG
1.203 brouard 2227: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2228: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2229: #endif
1.203 brouard 2230: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2231: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2232: #ifdef LINMINORIGINAL
1.188 brouard 2233: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2234: #else
2235: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2236: flatdir[i]=flat; /* Function is vanishing in that direction i */
2237: #endif
2238: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2239: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2240: /* because that direction will be replaced unless the gain del is small */
2241: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2242: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2243: /* with the new direction. */
2244: del=fabs(fptt-(*fret));
2245: ibig=i;
1.126 brouard 2246: }
2247: #ifdef DEBUG
2248: printf("%d %.12e",i,(*fret));
2249: fprintf(ficlog,"%d %.12e",i,(*fret));
2250: for (j=1;j<=n;j++) {
1.224 brouard 2251: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2252: printf(" x(%d)=%.12e",j,xit[j]);
2253: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2254: }
2255: for(j=1;j<=n;j++) {
1.225 brouard 2256: printf(" p(%d)=%.12e",j,p[j]);
2257: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2258: }
2259: printf("\n");
2260: fprintf(ficlog,"\n");
2261: #endif
1.187 brouard 2262: } /* end loop on each direction i */
2263: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2264: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2265: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2266: for(j=1;j<=n;j++) {
1.225 brouard 2267: if(flatdir[j] >0){
2268: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2269: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2270: }
2271: /* printf("\n"); */
2272: /* fprintf(ficlog,"\n"); */
2273: }
1.243 brouard 2274: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2275: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2276: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2277: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2278: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2279: /* decreased of more than 3.84 */
2280: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2281: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2282: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2283:
1.188 brouard 2284: /* Starting the program with initial values given by a former maximization will simply change */
2285: /* the scales of the directions and the directions, because the are reset to canonical directions */
2286: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2287: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2288: #ifdef DEBUG
2289: int k[2],l;
2290: k[0]=1;
2291: k[1]=-1;
2292: printf("Max: %.12e",(*func)(p));
2293: fprintf(ficlog,"Max: %.12e",(*func)(p));
2294: for (j=1;j<=n;j++) {
2295: printf(" %.12e",p[j]);
2296: fprintf(ficlog," %.12e",p[j]);
2297: }
2298: printf("\n");
2299: fprintf(ficlog,"\n");
2300: for(l=0;l<=1;l++) {
2301: for (j=1;j<=n;j++) {
2302: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2303: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2304: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2305: }
2306: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2307: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2308: }
2309: #endif
2310:
1.224 brouard 2311: #ifdef LINMINORIGINAL
2312: #else
2313: free_ivector(flatdir,1,n);
2314: #endif
1.126 brouard 2315: free_vector(xit,1,n);
2316: free_vector(xits,1,n);
2317: free_vector(ptt,1,n);
2318: free_vector(pt,1,n);
2319: return;
1.192 brouard 2320: } /* enough precision */
1.240 brouard 2321: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2322: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2323: ptt[j]=2.0*p[j]-pt[j];
2324: xit[j]=p[j]-pt[j];
2325: pt[j]=p[j];
2326: }
1.181 brouard 2327: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2328: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2329: if (*iter <=4) {
1.225 brouard 2330: #else
2331: #endif
1.224 brouard 2332: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2333: #else
1.161 brouard 2334: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2335: #endif
1.162 brouard 2336: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2337: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2338: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2339: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2340: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2341: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2342: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2343: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2344: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2345: /* Even if f3 <f1, directest can be negative and t >0 */
2346: /* mu² and del² are equal when f3=f1 */
2347: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2348: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2349: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2350: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2351: #ifdef NRCORIGINAL
2352: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2353: #else
2354: 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 2355: t= t- del*SQR(fp-fptt);
1.183 brouard 2356: #endif
1.202 brouard 2357: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2358: #ifdef DEBUG
1.181 brouard 2359: 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);
2360: 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 2361: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2362: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2363: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2364: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2365: 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);
2366: 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);
2367: #endif
1.183 brouard 2368: #ifdef POWELLORIGINAL
2369: if (t < 0.0) { /* Then we use it for new direction */
2370: #else
1.182 brouard 2371: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2372: 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 2373: 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 2374: 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 2375: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2376: }
1.181 brouard 2377: if (directest < 0.0) { /* Then we use it for new direction */
2378: #endif
1.191 brouard 2379: #ifdef DEBUGLINMIN
1.234 brouard 2380: printf("Before linmin in direction P%d-P0\n",n);
2381: for (j=1;j<=n;j++) {
2382: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2383: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2384: if(j % ncovmodel == 0){
2385: printf("\n");
2386: fprintf(ficlog,"\n");
2387: }
2388: }
1.224 brouard 2389: #endif
2390: #ifdef LINMINORIGINAL
1.234 brouard 2391: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2392: #else
1.234 brouard 2393: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2394: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2395: #endif
1.234 brouard 2396:
1.191 brouard 2397: #ifdef DEBUGLINMIN
1.234 brouard 2398: for (j=1;j<=n;j++) {
2399: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2400: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2401: if(j % ncovmodel == 0){
2402: printf("\n");
2403: fprintf(ficlog,"\n");
2404: }
2405: }
1.224 brouard 2406: #endif
1.234 brouard 2407: for (j=1;j<=n;j++) {
2408: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2409: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2410: }
1.224 brouard 2411: #ifdef LINMINORIGINAL
2412: #else
1.234 brouard 2413: for (j=1, flatd=0;j<=n;j++) {
2414: if(flatdir[j]>0)
2415: flatd++;
2416: }
2417: if(flatd >0){
1.255 brouard 2418: printf("%d flat directions: ",flatd);
2419: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2420: for (j=1;j<=n;j++) {
2421: if(flatdir[j]>0){
2422: printf("%d ",j);
2423: fprintf(ficlog,"%d ",j);
2424: }
2425: }
2426: printf("\n");
2427: fprintf(ficlog,"\n");
2428: }
1.191 brouard 2429: #endif
1.234 brouard 2430: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2431: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2432:
1.126 brouard 2433: #ifdef DEBUG
1.234 brouard 2434: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2435: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2436: for(j=1;j<=n;j++){
2437: printf(" %lf",xit[j]);
2438: fprintf(ficlog," %lf",xit[j]);
2439: }
2440: printf("\n");
2441: fprintf(ficlog,"\n");
1.126 brouard 2442: #endif
1.192 brouard 2443: } /* end of t or directest negative */
1.224 brouard 2444: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2445: #else
1.234 brouard 2446: } /* end if (fptt < fp) */
1.192 brouard 2447: #endif
1.225 brouard 2448: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2449: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2450: #else
1.224 brouard 2451: #endif
1.234 brouard 2452: } /* loop iteration */
1.126 brouard 2453: }
1.234 brouard 2454:
1.126 brouard 2455: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2456:
1.235 brouard 2457: 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 2458: {
1.235 brouard 2459: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2460: (and selected quantitative values in nres)
2461: by left multiplying the unit
1.234 brouard 2462: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2463: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2464: /* Wx is row vector: population in state 1, population in state 2, population dead */
2465: /* or prevalence in state 1, prevalence in state 2, 0 */
2466: /* newm is the matrix after multiplications, its rows are identical at a factor */
2467: /* Initial matrix pimij */
2468: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2469: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2470: /* 0, 0 , 1} */
2471: /*
2472: * and after some iteration: */
2473: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2474: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2475: /* 0, 0 , 1} */
2476: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2477: /* {0.51571254859325999, 0.4842874514067399, */
2478: /* 0.51326036147820708, 0.48673963852179264} */
2479: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2480:
1.126 brouard 2481: int i, ii,j,k;
1.209 brouard 2482: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2483: /* double **matprod2(); */ /* test */
1.218 brouard 2484: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2485: double **newm;
1.209 brouard 2486: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2487: int ncvloop=0;
1.169 brouard 2488:
1.209 brouard 2489: min=vector(1,nlstate);
2490: max=vector(1,nlstate);
2491: meandiff=vector(1,nlstate);
2492:
1.218 brouard 2493: /* Starting with matrix unity */
1.126 brouard 2494: for (ii=1;ii<=nlstate+ndeath;ii++)
2495: for (j=1;j<=nlstate+ndeath;j++){
2496: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2497: }
1.169 brouard 2498:
2499: cov[1]=1.;
2500:
2501: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2502: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2503: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2504: ncvloop++;
1.126 brouard 2505: newm=savm;
2506: /* Covariates have to be included here again */
1.138 brouard 2507: cov[2]=agefin;
1.187 brouard 2508: if(nagesqr==1)
2509: cov[3]= agefin*agefin;;
1.234 brouard 2510: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2511: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2512: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2513: /* 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 2514: }
2515: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2516: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2517: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2518: /* 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 2519: }
1.237 brouard 2520: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2521: if(Dummy[Tvar[Tage[k]]]){
2522: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2523: } else{
1.235 brouard 2524: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2525: }
1.235 brouard 2526: /* 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 2527: }
1.237 brouard 2528: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2529: /* 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 2530: if(Dummy[Tvard[k][1]==0]){
2531: if(Dummy[Tvard[k][2]==0]){
2532: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2533: }else{
2534: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2535: }
2536: }else{
2537: if(Dummy[Tvard[k][2]==0]){
2538: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2539: }else{
2540: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2541: }
2542: }
1.234 brouard 2543: }
1.138 brouard 2544: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2545: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2546: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2547: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2548: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2549: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2550: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2551:
1.126 brouard 2552: savm=oldm;
2553: oldm=newm;
1.209 brouard 2554:
2555: for(j=1; j<=nlstate; j++){
2556: max[j]=0.;
2557: min[j]=1.;
2558: }
2559: for(i=1;i<=nlstate;i++){
2560: sumnew=0;
2561: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2562: for(j=1; j<=nlstate; j++){
2563: prlim[i][j]= newm[i][j]/(1-sumnew);
2564: max[j]=FMAX(max[j],prlim[i][j]);
2565: min[j]=FMIN(min[j],prlim[i][j]);
2566: }
2567: }
2568:
1.126 brouard 2569: maxmax=0.;
1.209 brouard 2570: for(j=1; j<=nlstate; j++){
2571: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2572: maxmax=FMAX(maxmax,meandiff[j]);
2573: /* 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 2574: } /* j loop */
1.203 brouard 2575: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2576: /* 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 2577: if(maxmax < ftolpl){
1.209 brouard 2578: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2579: free_vector(min,1,nlstate);
2580: free_vector(max,1,nlstate);
2581: free_vector(meandiff,1,nlstate);
1.126 brouard 2582: return prlim;
2583: }
1.169 brouard 2584: } /* age loop */
1.208 brouard 2585: /* After some age loop it doesn't converge */
1.209 brouard 2586: 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 2587: 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 2588: /* 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); */
2589: free_vector(min,1,nlstate);
2590: free_vector(max,1,nlstate);
2591: free_vector(meandiff,1,nlstate);
1.208 brouard 2592:
1.169 brouard 2593: return prlim; /* should not reach here */
1.126 brouard 2594: }
2595:
1.217 brouard 2596:
2597: /**** Back Prevalence limit (stable or period prevalence) ****************/
2598:
1.218 brouard 2599: /* 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) */
2600: /* 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 2601: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2602: {
1.218 brouard 2603: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2604: matrix by transitions matrix until convergence is reached with precision ftolpl */
2605: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2606: /* Wx is row vector: population in state 1, population in state 2, population dead */
2607: /* or prevalence in state 1, prevalence in state 2, 0 */
2608: /* newm is the matrix after multiplications, its rows are identical at a factor */
2609: /* Initial matrix pimij */
2610: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2611: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2612: /* 0, 0 , 1} */
2613: /*
2614: * and after some iteration: */
2615: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2616: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2617: /* 0, 0 , 1} */
2618: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2619: /* {0.51571254859325999, 0.4842874514067399, */
2620: /* 0.51326036147820708, 0.48673963852179264} */
2621: /* If we start from prlim again, prlim tends to a constant matrix */
2622:
2623: int i, ii,j,k;
1.247 brouard 2624: int first=0;
1.217 brouard 2625: double *min, *max, *meandiff, maxmax,sumnew=0.;
2626: /* double **matprod2(); */ /* test */
2627: double **out, cov[NCOVMAX+1], **bmij();
2628: double **newm;
1.218 brouard 2629: double **dnewm, **doldm, **dsavm; /* for use */
2630: double **oldm, **savm; /* for use */
2631:
1.217 brouard 2632: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2633: int ncvloop=0;
2634:
2635: min=vector(1,nlstate);
2636: max=vector(1,nlstate);
2637: meandiff=vector(1,nlstate);
2638:
1.218 brouard 2639: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2640: oldm=oldms; savm=savms;
2641:
2642: /* Starting with matrix unity */
2643: for (ii=1;ii<=nlstate+ndeath;ii++)
2644: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2645: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2646: }
2647:
2648: cov[1]=1.;
2649:
2650: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2651: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2652: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2653: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2654: ncvloop++;
1.218 brouard 2655: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2656: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2657: /* Covariates have to be included here again */
2658: cov[2]=agefin;
2659: if(nagesqr==1)
2660: cov[3]= agefin*agefin;;
1.242 brouard 2661: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2662: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2663: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2664: /* printf("bprevalim Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
2665: }
2666: /* for (k=1; k<=cptcovn;k++) { */
2667: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2668: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2669: /* /\* 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])]); *\/ */
2670: /* } */
2671: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2672: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2673: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2674: /* 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]); */
2675: }
2676: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2677: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2678: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2679: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2680: for (k=1; k<=cptcovage;k++){ /* For product with age */
2681: if(Dummy[Tvar[Tage[k]]]){
2682: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2683: } else{
2684: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2685: }
2686: /* 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]); */
2687: }
2688: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2689: /* 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]); */
2690: if(Dummy[Tvard[k][1]==0]){
2691: if(Dummy[Tvard[k][2]==0]){
2692: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2693: }else{
2694: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2695: }
2696: }else{
2697: if(Dummy[Tvard[k][2]==0]){
2698: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2699: }else{
2700: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2701: }
2702: }
1.217 brouard 2703: }
2704:
2705: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2706: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2707: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2708: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2709: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2710: /* ij should be linked to the correct index of cov */
2711: /* age and covariate values ij are in 'cov', but we need to pass
2712: * ij for the observed prevalence at age and status and covariate
2713: * number: prevacurrent[(int)agefin][ii][ij]
2714: */
2715: /* 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 *\/ */
2716: /* 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 *\/ */
2717: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.217 brouard 2718: savm=oldm;
2719: oldm=newm;
2720: for(j=1; j<=nlstate; j++){
2721: max[j]=0.;
2722: min[j]=1.;
2723: }
2724: for(j=1; j<=nlstate; j++){
2725: for(i=1;i<=nlstate;i++){
1.234 brouard 2726: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2727: bprlim[i][j]= newm[i][j];
2728: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2729: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2730: }
2731: }
1.218 brouard 2732:
1.217 brouard 2733: maxmax=0.;
2734: for(i=1; i<=nlstate; i++){
2735: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2736: maxmax=FMAX(maxmax,meandiff[i]);
2737: /* 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); */
2738: } /* j loop */
2739: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2740: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2741: if(maxmax < ftolpl){
1.220 brouard 2742: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2743: free_vector(min,1,nlstate);
2744: free_vector(max,1,nlstate);
2745: free_vector(meandiff,1,nlstate);
2746: return bprlim;
2747: }
2748: } /* age loop */
2749: /* After some age loop it doesn't converge */
1.247 brouard 2750: if(first){
2751: first=1;
2752: 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\
2753: 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);
2754: }
2755: 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 2756: 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);
2757: /* 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); */
2758: free_vector(min,1,nlstate);
2759: free_vector(max,1,nlstate);
2760: free_vector(meandiff,1,nlstate);
2761:
2762: return bprlim; /* should not reach here */
2763: }
2764:
1.126 brouard 2765: /*************** transition probabilities ***************/
2766:
2767: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2768: {
1.138 brouard 2769: /* According to parameters values stored in x and the covariate's values stored in cov,
2770: computes the probability to be observed in state j being in state i by appying the
2771: model to the ncovmodel covariates (including constant and age).
2772: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2773: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2774: ncth covariate in the global vector x is given by the formula:
2775: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2776: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2777: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2778: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2779: Outputs ps[i][j] the probability to be observed in j being in j according to
2780: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2781: */
2782: double s1, lnpijopii;
1.126 brouard 2783: /*double t34;*/
1.164 brouard 2784: int i,j, nc, ii, jj;
1.126 brouard 2785:
1.223 brouard 2786: for(i=1; i<= nlstate; i++){
2787: for(j=1; j<i;j++){
2788: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2789: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2790: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2791: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2792: }
2793: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2794: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2795: }
2796: for(j=i+1; j<=nlstate+ndeath;j++){
2797: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2798: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2799: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2800: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2801: }
2802: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2803: }
2804: }
1.218 brouard 2805:
1.223 brouard 2806: for(i=1; i<= nlstate; i++){
2807: s1=0;
2808: for(j=1; j<i; j++){
2809: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2810: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2811: }
2812: for(j=i+1; j<=nlstate+ndeath; j++){
2813: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2814: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2815: }
2816: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2817: ps[i][i]=1./(s1+1.);
2818: /* Computing other pijs */
2819: for(j=1; j<i; j++)
2820: ps[i][j]= exp(ps[i][j])*ps[i][i];
2821: for(j=i+1; j<=nlstate+ndeath; j++)
2822: ps[i][j]= exp(ps[i][j])*ps[i][i];
2823: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2824: } /* end i */
1.218 brouard 2825:
1.223 brouard 2826: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2827: for(jj=1; jj<= nlstate+ndeath; jj++){
2828: ps[ii][jj]=0;
2829: ps[ii][ii]=1;
2830: }
2831: }
1.218 brouard 2832:
2833:
1.223 brouard 2834: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2835: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2836: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2837: /* } */
2838: /* printf("\n "); */
2839: /* } */
2840: /* printf("\n ");printf("%lf ",cov[2]);*/
2841: /*
2842: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2843: goto end;*/
1.223 brouard 2844: return ps;
1.126 brouard 2845: }
2846:
1.218 brouard 2847: /*************** backward transition probabilities ***************/
2848:
2849: /* 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 ) */
2850: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2851: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2852: {
1.222 brouard 2853: /* Computes the backward probability at age agefin and covariate ij
2854: * and returns in **ps as well as **bmij.
2855: */
1.218 brouard 2856: int i, ii, j,k;
1.222 brouard 2857:
2858: double **out, **pmij();
2859: double sumnew=0.;
1.218 brouard 2860: double agefin;
1.222 brouard 2861:
2862: double **dnewm, **dsavm, **doldm;
2863: double **bbmij;
2864:
1.218 brouard 2865: doldm=ddoldms; /* global pointers */
1.222 brouard 2866: dnewm=ddnewms;
2867: dsavm=ddsavms;
2868:
2869: agefin=cov[2];
2870: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2871: the observed prevalence (with this covariate ij) */
2872: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2873: /* We do have the matrix Px in savm and we need pij */
2874: for (j=1;j<=nlstate+ndeath;j++){
2875: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2876: for (ii=1;ii<=nlstate;ii++){
2877: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2878: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2879: for (ii=1;ii<=nlstate+ndeath;ii++){
2880: if(sumnew >= 1.e-10){
2881: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2882: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2883: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2884: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2885: /* }else */
2886: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2887: }else{
1.242 brouard 2888: ;
2889: /* printf("ii=%d, i=%d, doldm=%lf dsavm=%lf, probs=%lf, sumnew=%lf,agefin=%d\n",ii,j,doldm[ii][j],dsavm[ii][j],prevacurrent[(int)agefin][ii][ij],sumnew, (int)agefin); */
1.222 brouard 2890: }
2891: } /*End ii */
2892: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2893: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2894: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2895: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2896: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2897: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2898: /* left Product of this matrix by diag matrix of prevalences (savm) */
2899: for (j=1;j<=nlstate+ndeath;j++){
2900: for (ii=1;ii<=nlstate+ndeath;ii++){
2901: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2902: }
2903: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2904: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2905: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2906: /* end bmij */
2907: return ps;
1.218 brouard 2908: }
1.217 brouard 2909: /*************** transition probabilities ***************/
2910:
1.218 brouard 2911: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2912: {
2913: /* According to parameters values stored in x and the covariate's values stored in cov,
2914: computes the probability to be observed in state j being in state i by appying the
2915: model to the ncovmodel covariates (including constant and age).
2916: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2917: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2918: ncth covariate in the global vector x is given by the formula:
2919: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2920: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2921: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2922: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2923: Outputs ps[i][j] the probability to be observed in j being in j according to
2924: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2925: */
2926: double s1, lnpijopii;
2927: /*double t34;*/
2928: int i,j, nc, ii, jj;
2929:
1.234 brouard 2930: for(i=1; i<= nlstate; i++){
2931: for(j=1; j<i;j++){
2932: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2933: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2934: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2935: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2936: }
2937: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2938: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2939: }
2940: for(j=i+1; j<=nlstate+ndeath;j++){
2941: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2942: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2943: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2944: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2945: }
2946: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2947: }
2948: }
2949:
2950: for(i=1; i<= nlstate; i++){
2951: s1=0;
2952: for(j=1; j<i; j++){
2953: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2954: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2955: }
2956: for(j=i+1; j<=nlstate+ndeath; j++){
2957: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2958: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2959: }
2960: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2961: ps[i][i]=1./(s1+1.);
2962: /* Computing other pijs */
2963: for(j=1; j<i; j++)
2964: ps[i][j]= exp(ps[i][j])*ps[i][i];
2965: for(j=i+1; j<=nlstate+ndeath; j++)
2966: ps[i][j]= exp(ps[i][j])*ps[i][i];
2967: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2968: } /* end i */
2969:
2970: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2971: for(jj=1; jj<= nlstate+ndeath; jj++){
2972: ps[ii][jj]=0;
2973: ps[ii][ii]=1;
2974: }
2975: }
2976: /* Added for backcast */ /* Transposed matrix too */
2977: for(jj=1; jj<= nlstate+ndeath; jj++){
2978: s1=0.;
2979: for(ii=1; ii<= nlstate+ndeath; ii++){
2980: s1+=ps[ii][jj];
2981: }
2982: for(ii=1; ii<= nlstate; ii++){
2983: ps[ii][jj]=ps[ii][jj]/s1;
2984: }
2985: }
2986: /* Transposition */
2987: for(jj=1; jj<= nlstate+ndeath; jj++){
2988: for(ii=jj; ii<= nlstate+ndeath; ii++){
2989: s1=ps[ii][jj];
2990: ps[ii][jj]=ps[jj][ii];
2991: ps[jj][ii]=s1;
2992: }
2993: }
2994: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2995: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2996: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2997: /* } */
2998: /* printf("\n "); */
2999: /* } */
3000: /* printf("\n ");printf("%lf ",cov[2]);*/
3001: /*
3002: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3003: goto end;*/
3004: return ps;
1.217 brouard 3005: }
3006:
3007:
1.126 brouard 3008: /**************** Product of 2 matrices ******************/
3009:
1.145 brouard 3010: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3011: {
3012: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3013: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3014: /* in, b, out are matrice of pointers which should have been initialized
3015: before: only the contents of out is modified. The function returns
3016: a pointer to pointers identical to out */
1.145 brouard 3017: int i, j, k;
1.126 brouard 3018: for(i=nrl; i<= nrh; i++)
1.145 brouard 3019: for(k=ncolol; k<=ncoloh; k++){
3020: out[i][k]=0.;
3021: for(j=ncl; j<=nch; j++)
3022: out[i][k] +=in[i][j]*b[j][k];
3023: }
1.126 brouard 3024: return out;
3025: }
3026:
3027:
3028: /************* Higher Matrix Product ***************/
3029:
1.235 brouard 3030: 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 3031: {
1.218 brouard 3032: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3033: 'nhstepm*hstepm*stepm' months (i.e. until
3034: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3035: nhstepm*hstepm matrices.
3036: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3037: (typically every 2 years instead of every month which is too big
3038: for the memory).
3039: Model is determined by parameters x and covariates have to be
3040: included manually here.
3041:
3042: */
3043:
3044: int i, j, d, h, k;
1.131 brouard 3045: double **out, cov[NCOVMAX+1];
1.126 brouard 3046: double **newm;
1.187 brouard 3047: double agexact;
1.214 brouard 3048: double agebegin, ageend;
1.126 brouard 3049:
3050: /* Hstepm could be zero and should return the unit matrix */
3051: for (i=1;i<=nlstate+ndeath;i++)
3052: for (j=1;j<=nlstate+ndeath;j++){
3053: oldm[i][j]=(i==j ? 1.0 : 0.0);
3054: po[i][j][0]=(i==j ? 1.0 : 0.0);
3055: }
3056: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3057: for(h=1; h <=nhstepm; h++){
3058: for(d=1; d <=hstepm; d++){
3059: newm=savm;
3060: /* Covariates have to be included here again */
3061: cov[1]=1.;
1.214 brouard 3062: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3063: cov[2]=agexact;
3064: if(nagesqr==1)
1.227 brouard 3065: cov[3]= agexact*agexact;
1.235 brouard 3066: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3067: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3068: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3069: /* 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)); */
3070: }
3071: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3072: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3073: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3074: /* 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]); */
3075: }
3076: for (k=1; k<=cptcovage;k++){
3077: if(Dummy[Tvar[Tage[k]]]){
3078: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3079: } else{
3080: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3081: }
3082: /* 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]); */
3083: }
3084: for (k=1; k<=cptcovprod;k++){ /* */
3085: /* 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]); */
3086: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3087: }
3088: /* for (k=1; k<=cptcovn;k++) */
3089: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3090: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3091: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3092: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3093: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3094:
3095:
1.126 brouard 3096: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3097: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3098: /* right multiplication of oldm by the current matrix */
1.126 brouard 3099: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3100: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3101: /* if((int)age == 70){ */
3102: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3103: /* for(i=1; i<=nlstate+ndeath; i++) { */
3104: /* printf("%d pmmij ",i); */
3105: /* for(j=1;j<=nlstate+ndeath;j++) { */
3106: /* printf("%f ",pmmij[i][j]); */
3107: /* } */
3108: /* printf(" oldm "); */
3109: /* for(j=1;j<=nlstate+ndeath;j++) { */
3110: /* printf("%f ",oldm[i][j]); */
3111: /* } */
3112: /* printf("\n"); */
3113: /* } */
3114: /* } */
1.126 brouard 3115: savm=oldm;
3116: oldm=newm;
3117: }
3118: for(i=1; i<=nlstate+ndeath; i++)
3119: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3120: po[i][j][h]=newm[i][j];
3121: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3122: }
1.128 brouard 3123: /*printf("h=%d ",h);*/
1.126 brouard 3124: } /* end h */
1.218 brouard 3125: /* printf("\n H=%d \n",h); */
1.126 brouard 3126: return po;
3127: }
3128:
1.217 brouard 3129: /************* Higher Back Matrix Product ***************/
1.218 brouard 3130: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.222 brouard 3131: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3132: {
1.218 brouard 3133: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3134: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3135: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3136: nhstepm*hstepm matrices.
3137: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3138: (typically every 2 years instead of every month which is too big
1.217 brouard 3139: for the memory).
1.218 brouard 3140: Model is determined by parameters x and covariates have to be
3141: included manually here.
1.217 brouard 3142:
1.222 brouard 3143: */
1.217 brouard 3144:
3145: int i, j, d, h, k;
3146: double **out, cov[NCOVMAX+1];
3147: double **newm;
3148: double agexact;
3149: double agebegin, ageend;
1.222 brouard 3150: double **oldm, **savm;
1.217 brouard 3151:
1.222 brouard 3152: oldm=oldms;savm=savms;
1.217 brouard 3153: /* Hstepm could be zero and should return the unit matrix */
3154: for (i=1;i<=nlstate+ndeath;i++)
3155: for (j=1;j<=nlstate+ndeath;j++){
3156: oldm[i][j]=(i==j ? 1.0 : 0.0);
3157: po[i][j][0]=(i==j ? 1.0 : 0.0);
3158: }
3159: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3160: for(h=1; h <=nhstepm; h++){
3161: for(d=1; d <=hstepm; d++){
3162: newm=savm;
3163: /* Covariates have to be included here again */
3164: cov[1]=1.;
3165: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3166: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3167: cov[2]=agexact;
3168: if(nagesqr==1)
1.222 brouard 3169: cov[3]= agexact*agexact;
1.218 brouard 3170: for (k=1; k<=cptcovn;k++)
1.222 brouard 3171: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3172: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3173: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3174: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3175: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3176: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3177: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3178: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3179: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
1.218 brouard 3180:
3181:
1.217 brouard 3182: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3183: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3184: /* Careful transposed matrix */
1.222 brouard 3185: /* age is in cov[2] */
1.218 brouard 3186: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3187: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3188: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3189: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3190: /* if((int)age == 70){ */
3191: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3192: /* for(i=1; i<=nlstate+ndeath; i++) { */
3193: /* printf("%d pmmij ",i); */
3194: /* for(j=1;j<=nlstate+ndeath;j++) { */
3195: /* printf("%f ",pmmij[i][j]); */
3196: /* } */
3197: /* printf(" oldm "); */
3198: /* for(j=1;j<=nlstate+ndeath;j++) { */
3199: /* printf("%f ",oldm[i][j]); */
3200: /* } */
3201: /* printf("\n"); */
3202: /* } */
3203: /* } */
3204: savm=oldm;
3205: oldm=newm;
3206: }
3207: for(i=1; i<=nlstate+ndeath; i++)
3208: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3209: po[i][j][h]=newm[i][j];
3210: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3211: }
3212: /*printf("h=%d ",h);*/
3213: } /* end h */
1.222 brouard 3214: /* printf("\n H=%d \n",h); */
1.217 brouard 3215: return po;
3216: }
3217:
3218:
1.162 brouard 3219: #ifdef NLOPT
3220: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3221: double fret;
3222: double *xt;
3223: int j;
3224: myfunc_data *d2 = (myfunc_data *) pd;
3225: /* xt = (p1-1); */
3226: xt=vector(1,n);
3227: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3228:
3229: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3230: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3231: printf("Function = %.12lf ",fret);
3232: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3233: printf("\n");
3234: free_vector(xt,1,n);
3235: return fret;
3236: }
3237: #endif
1.126 brouard 3238:
3239: /*************** log-likelihood *************/
3240: double func( double *x)
3241: {
1.226 brouard 3242: int i, ii, j, k, mi, d, kk;
3243: int ioffset=0;
3244: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3245: double **out;
3246: double lli; /* Individual log likelihood */
3247: int s1, s2;
1.228 brouard 3248: 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 3249: double bbh, survp;
3250: long ipmx;
3251: double agexact;
3252: /*extern weight */
3253: /* We are differentiating ll according to initial status */
3254: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3255: /*for(i=1;i<imx;i++)
3256: printf(" %d\n",s[4][i]);
3257: */
1.162 brouard 3258:
1.226 brouard 3259: ++countcallfunc;
1.162 brouard 3260:
1.226 brouard 3261: cov[1]=1.;
1.126 brouard 3262:
1.226 brouard 3263: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3264: ioffset=0;
1.226 brouard 3265: if(mle==1){
3266: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3267: /* Computes the values of the ncovmodel covariates of the model
3268: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3269: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3270: to be observed in j being in i according to the model.
3271: */
1.243 brouard 3272: ioffset=2+nagesqr ;
1.233 brouard 3273: /* Fixed */
1.234 brouard 3274: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3275: 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)*/
3276: }
1.226 brouard 3277: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3278: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3279: has been calculated etc */
3280: /* For an individual i, wav[i] gives the number of effective waves */
3281: /* We compute the contribution to Likelihood of each effective transition
3282: mw[mi][i] is real wave of the mi th effectve wave */
3283: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3284: s2=s[mw[mi+1][i]][i];
3285: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3286: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3287: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3288: */
3289: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3290: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3291: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3292: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3293: }
3294: for (ii=1;ii<=nlstate+ndeath;ii++)
3295: for (j=1;j<=nlstate+ndeath;j++){
3296: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3297: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3298: }
3299: for(d=0; d<dh[mi][i]; d++){
3300: newm=savm;
3301: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3302: cov[2]=agexact;
3303: if(nagesqr==1)
3304: cov[3]= agexact*agexact; /* Should be changed here */
3305: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3306: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3307: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3308: else
3309: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3310: }
3311: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3312: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3313: savm=oldm;
3314: oldm=newm;
3315: } /* end mult */
3316:
3317: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3318: /* But now since version 0.9 we anticipate for bias at large stepm.
3319: * If stepm is larger than one month (smallest stepm) and if the exact delay
3320: * (in months) between two waves is not a multiple of stepm, we rounded to
3321: * the nearest (and in case of equal distance, to the lowest) interval but now
3322: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3323: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3324: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3325: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3326: * -stepm/2 to stepm/2 .
3327: * For stepm=1 the results are the same as for previous versions of Imach.
3328: * For stepm > 1 the results are less biased than in previous versions.
3329: */
1.234 brouard 3330: s1=s[mw[mi][i]][i];
3331: s2=s[mw[mi+1][i]][i];
3332: bbh=(double)bh[mi][i]/(double)stepm;
3333: /* bias bh is positive if real duration
3334: * is higher than the multiple of stepm and negative otherwise.
3335: */
3336: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3337: if( s2 > nlstate){
3338: /* i.e. if s2 is a death state and if the date of death is known
3339: then the contribution to the likelihood is the probability to
3340: die between last step unit time and current step unit time,
3341: which is also equal to probability to die before dh
3342: minus probability to die before dh-stepm .
3343: In version up to 0.92 likelihood was computed
3344: as if date of death was unknown. Death was treated as any other
3345: health state: the date of the interview describes the actual state
3346: and not the date of a change in health state. The former idea was
3347: to consider that at each interview the state was recorded
3348: (healthy, disable or death) and IMaCh was corrected; but when we
3349: introduced the exact date of death then we should have modified
3350: the contribution of an exact death to the likelihood. This new
3351: contribution is smaller and very dependent of the step unit
3352: stepm. It is no more the probability to die between last interview
3353: and month of death but the probability to survive from last
3354: interview up to one month before death multiplied by the
3355: probability to die within a month. Thanks to Chris
3356: Jackson for correcting this bug. Former versions increased
3357: mortality artificially. The bad side is that we add another loop
3358: which slows down the processing. The difference can be up to 10%
3359: lower mortality.
3360: */
3361: /* If, at the beginning of the maximization mostly, the
3362: cumulative probability or probability to be dead is
3363: constant (ie = 1) over time d, the difference is equal to
3364: 0. out[s1][3] = savm[s1][3]: probability, being at state
3365: s1 at precedent wave, to be dead a month before current
3366: wave is equal to probability, being at state s1 at
3367: precedent wave, to be dead at mont of the current
3368: wave. Then the observed probability (that this person died)
3369: is null according to current estimated parameter. In fact,
3370: it should be very low but not zero otherwise the log go to
3371: infinity.
3372: */
1.183 brouard 3373: /* #ifdef INFINITYORIGINAL */
3374: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3375: /* #else */
3376: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3377: /* lli=log(mytinydouble); */
3378: /* else */
3379: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3380: /* #endif */
1.226 brouard 3381: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3382:
1.226 brouard 3383: } else if ( s2==-1 ) { /* alive */
3384: for (j=1,survp=0. ; j<=nlstate; j++)
3385: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3386: /*survp += out[s1][j]; */
3387: lli= log(survp);
3388: }
3389: else if (s2==-4) {
3390: for (j=3,survp=0. ; j<=nlstate; j++)
3391: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3392: lli= log(survp);
3393: }
3394: else if (s2==-5) {
3395: for (j=1,survp=0. ; j<=2; j++)
3396: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3397: lli= log(survp);
3398: }
3399: else{
3400: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3401: /* 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 */
3402: }
3403: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3404: /*if(lli ==000.0)*/
3405: /*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); */
3406: ipmx +=1;
3407: sw += weight[i];
3408: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3409: /* if (lli < log(mytinydouble)){ */
3410: /* 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); */
3411: /* 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]); */
3412: /* } */
3413: } /* end of wave */
3414: } /* end of individual */
3415: } else if(mle==2){
3416: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3417: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3418: for(mi=1; mi<= wav[i]-1; mi++){
3419: for (ii=1;ii<=nlstate+ndeath;ii++)
3420: for (j=1;j<=nlstate+ndeath;j++){
3421: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3422: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3423: }
3424: for(d=0; d<=dh[mi][i]; d++){
3425: newm=savm;
3426: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3427: cov[2]=agexact;
3428: if(nagesqr==1)
3429: cov[3]= agexact*agexact;
3430: for (kk=1; kk<=cptcovage;kk++) {
3431: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3432: }
3433: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3434: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3435: savm=oldm;
3436: oldm=newm;
3437: } /* end mult */
3438:
3439: s1=s[mw[mi][i]][i];
3440: s2=s[mw[mi+1][i]][i];
3441: bbh=(double)bh[mi][i]/(double)stepm;
3442: 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 */
3443: ipmx +=1;
3444: sw += weight[i];
3445: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3446: } /* end of wave */
3447: } /* end of individual */
3448: } else if(mle==3){ /* exponential inter-extrapolation */
3449: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3450: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3451: for(mi=1; mi<= wav[i]-1; mi++){
3452: for (ii=1;ii<=nlstate+ndeath;ii++)
3453: for (j=1;j<=nlstate+ndeath;j++){
3454: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3455: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3456: }
3457: for(d=0; d<dh[mi][i]; d++){
3458: newm=savm;
3459: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3460: cov[2]=agexact;
3461: if(nagesqr==1)
3462: cov[3]= agexact*agexact;
3463: for (kk=1; kk<=cptcovage;kk++) {
3464: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3465: }
3466: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3467: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3468: savm=oldm;
3469: oldm=newm;
3470: } /* end mult */
3471:
3472: s1=s[mw[mi][i]][i];
3473: s2=s[mw[mi+1][i]][i];
3474: bbh=(double)bh[mi][i]/(double)stepm;
3475: 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 */
3476: ipmx +=1;
3477: sw += weight[i];
3478: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3479: } /* end of wave */
3480: } /* end of individual */
3481: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3482: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3483: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3484: for(mi=1; mi<= wav[i]-1; mi++){
3485: for (ii=1;ii<=nlstate+ndeath;ii++)
3486: for (j=1;j<=nlstate+ndeath;j++){
3487: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3488: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3489: }
3490: for(d=0; d<dh[mi][i]; d++){
3491: newm=savm;
3492: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3493: cov[2]=agexact;
3494: if(nagesqr==1)
3495: cov[3]= agexact*agexact;
3496: for (kk=1; kk<=cptcovage;kk++) {
3497: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3498: }
1.126 brouard 3499:
1.226 brouard 3500: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3501: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3502: savm=oldm;
3503: oldm=newm;
3504: } /* end mult */
3505:
3506: s1=s[mw[mi][i]][i];
3507: s2=s[mw[mi+1][i]][i];
3508: if( s2 > nlstate){
3509: lli=log(out[s1][s2] - savm[s1][s2]);
3510: } else if ( s2==-1 ) { /* alive */
3511: for (j=1,survp=0. ; j<=nlstate; j++)
3512: survp += out[s1][j];
3513: lli= log(survp);
3514: }else{
3515: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3516: }
3517: ipmx +=1;
3518: sw += weight[i];
3519: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3520: /* 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 3521: } /* end of wave */
3522: } /* end of individual */
3523: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3524: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3525: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3526: for(mi=1; mi<= wav[i]-1; mi++){
3527: for (ii=1;ii<=nlstate+ndeath;ii++)
3528: for (j=1;j<=nlstate+ndeath;j++){
3529: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3530: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3531: }
3532: for(d=0; d<dh[mi][i]; d++){
3533: newm=savm;
3534: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3535: cov[2]=agexact;
3536: if(nagesqr==1)
3537: cov[3]= agexact*agexact;
3538: for (kk=1; kk<=cptcovage;kk++) {
3539: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3540: }
1.126 brouard 3541:
1.226 brouard 3542: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3543: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3544: savm=oldm;
3545: oldm=newm;
3546: } /* end mult */
3547:
3548: s1=s[mw[mi][i]][i];
3549: s2=s[mw[mi+1][i]][i];
3550: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3551: ipmx +=1;
3552: sw += weight[i];
3553: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3554: /*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]);*/
3555: } /* end of wave */
3556: } /* end of individual */
3557: } /* End of if */
3558: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3559: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3560: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3561: return -l;
1.126 brouard 3562: }
3563:
3564: /*************** log-likelihood *************/
3565: double funcone( double *x)
3566: {
1.228 brouard 3567: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3568: int i, ii, j, k, mi, d, kk;
1.228 brouard 3569: int ioffset=0;
1.131 brouard 3570: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3571: double **out;
3572: double lli; /* Individual log likelihood */
3573: double llt;
3574: int s1, s2;
1.228 brouard 3575: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3576:
1.126 brouard 3577: double bbh, survp;
1.187 brouard 3578: double agexact;
1.214 brouard 3579: double agebegin, ageend;
1.126 brouard 3580: /*extern weight */
3581: /* We are differentiating ll according to initial status */
3582: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3583: /*for(i=1;i<imx;i++)
3584: printf(" %d\n",s[4][i]);
3585: */
3586: cov[1]=1.;
3587:
3588: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3589: ioffset=0;
3590: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3591: /* ioffset=2+nagesqr+cptcovage; */
3592: ioffset=2+nagesqr;
1.232 brouard 3593: /* Fixed */
1.224 brouard 3594: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3595: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3596: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3597: 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)*/
3598: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3599: /* cov[2+6]=covar[Tvar[6]][i]; */
3600: /* cov[2+6]=covar[2][i]; V2 */
3601: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3602: /* cov[2+7]=covar[Tvar[7]][i]; */
3603: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3604: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3605: /* cov[2+9]=covar[Tvar[9]][i]; */
3606: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3607: }
1.232 brouard 3608: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3609: /* 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?)*\/ */
3610: /* } */
1.231 brouard 3611: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3612: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3613: /* } */
1.225 brouard 3614:
1.233 brouard 3615:
3616: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3617: /* Wave varying (but not age varying) */
3618: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3619: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3620: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3621: }
1.232 brouard 3622: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3623: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3624: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3625: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3626: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3627: /* 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 3628: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3629: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3630: /* /\* 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]); *\/ */
3631: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3632: /* } */
1.126 brouard 3633: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3634: for (j=1;j<=nlstate+ndeath;j++){
3635: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3636: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3637: }
1.214 brouard 3638:
3639: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3640: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3641: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3642: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3643: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3644: and mw[mi+1][i]. dh depends on stepm.*/
3645: newm=savm;
1.247 brouard 3646: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3647: cov[2]=agexact;
3648: if(nagesqr==1)
3649: cov[3]= agexact*agexact;
3650: for (kk=1; kk<=cptcovage;kk++) {
3651: if(!FixedV[Tvar[Tage[kk]]])
3652: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3653: else
3654: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3655: }
3656: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3657: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3658: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3659: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3660: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3661: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3662: savm=oldm;
3663: oldm=newm;
1.126 brouard 3664: } /* end mult */
3665:
3666: s1=s[mw[mi][i]][i];
3667: s2=s[mw[mi+1][i]][i];
1.217 brouard 3668: /* if(s2==-1){ */
3669: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3670: /* /\* exit(1); *\/ */
3671: /* } */
1.126 brouard 3672: bbh=(double)bh[mi][i]/(double)stepm;
3673: /* bias is positive if real duration
3674: * is higher than the multiple of stepm and negative otherwise.
3675: */
3676: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3677: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3678: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3679: for (j=1,survp=0. ; j<=nlstate; j++)
3680: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3681: lli= log(survp);
1.126 brouard 3682: }else if (mle==1){
1.242 brouard 3683: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3684: } else if(mle==2){
1.242 brouard 3685: 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 3686: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3687: 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 3688: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3689: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3690: } else{ /* mle=0 back to 1 */
1.242 brouard 3691: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3692: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3693: } /* End of if */
3694: ipmx +=1;
3695: sw += weight[i];
3696: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3697: /*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 3698: if(globpr){
1.246 brouard 3699: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3700: %11.6f %11.6f %11.6f ", \
1.242 brouard 3701: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3702: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3703: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3704: llt +=ll[k]*gipmx/gsw;
3705: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3706: }
3707: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3708: }
1.232 brouard 3709: } /* end of wave */
3710: } /* end of individual */
3711: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3712: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3713: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3714: if(globpr==0){ /* First time we count the contributions and weights */
3715: gipmx=ipmx;
3716: gsw=sw;
3717: }
3718: return -l;
1.126 brouard 3719: }
3720:
3721:
3722: /*************** function likelione ***********/
3723: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3724: {
3725: /* This routine should help understanding what is done with
3726: the selection of individuals/waves and
3727: to check the exact contribution to the likelihood.
3728: Plotting could be done.
3729: */
3730: int k;
3731:
3732: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3733: strcpy(fileresilk,"ILK_");
1.202 brouard 3734: strcat(fileresilk,fileresu);
1.126 brouard 3735: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3736: printf("Problem with resultfile: %s\n", fileresilk);
3737: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3738: }
1.214 brouard 3739: 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");
3740: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3741: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3742: for(k=1; k<=nlstate; k++)
3743: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3744: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3745: }
3746:
3747: *fretone=(*funcone)(p);
3748: if(*globpri !=0){
3749: fclose(ficresilk);
1.205 brouard 3750: if (mle ==0)
3751: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3752: else if(mle >=1)
3753: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3754: 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 3755:
1.208 brouard 3756:
3757: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3758: 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 3759: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3760: }
1.207 brouard 3761: 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 3762: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3763: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3764: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3765: fflush(fichtm);
1.205 brouard 3766: }
1.126 brouard 3767: return;
3768: }
3769:
3770:
3771: /*********** Maximum Likelihood Estimation ***************/
3772:
3773: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3774: {
1.165 brouard 3775: int i,j, iter=0;
1.126 brouard 3776: double **xi;
3777: double fret;
3778: double fretone; /* Only one call to likelihood */
3779: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3780:
3781: #ifdef NLOPT
3782: int creturn;
3783: nlopt_opt opt;
3784: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3785: double *lb;
3786: double minf; /* the minimum objective value, upon return */
3787: double * p1; /* Shifted parameters from 0 instead of 1 */
3788: myfunc_data dinst, *d = &dinst;
3789: #endif
3790:
3791:
1.126 brouard 3792: xi=matrix(1,npar,1,npar);
3793: for (i=1;i<=npar;i++)
3794: for (j=1;j<=npar;j++)
3795: xi[i][j]=(i==j ? 1.0 : 0.0);
3796: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3797: strcpy(filerespow,"POW_");
1.126 brouard 3798: strcat(filerespow,fileres);
3799: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3800: printf("Problem with resultfile: %s\n", filerespow);
3801: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3802: }
3803: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3804: for (i=1;i<=nlstate;i++)
3805: for(j=1;j<=nlstate+ndeath;j++)
3806: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3807: fprintf(ficrespow,"\n");
1.162 brouard 3808: #ifdef POWELL
1.126 brouard 3809: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3810: #endif
1.126 brouard 3811:
1.162 brouard 3812: #ifdef NLOPT
3813: #ifdef NEWUOA
3814: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3815: #else
3816: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3817: #endif
3818: lb=vector(0,npar-1);
3819: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3820: nlopt_set_lower_bounds(opt, lb);
3821: nlopt_set_initial_step1(opt, 0.1);
3822:
3823: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3824: d->function = func;
3825: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3826: nlopt_set_min_objective(opt, myfunc, d);
3827: nlopt_set_xtol_rel(opt, ftol);
3828: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3829: printf("nlopt failed! %d\n",creturn);
3830: }
3831: else {
3832: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3833: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3834: iter=1; /* not equal */
3835: }
3836: nlopt_destroy(opt);
3837: #endif
1.126 brouard 3838: free_matrix(xi,1,npar,1,npar);
3839: fclose(ficrespow);
1.203 brouard 3840: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3841: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3842: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3843:
3844: }
3845:
3846: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3847: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3848: {
3849: double **a,**y,*x,pd;
1.203 brouard 3850: /* double **hess; */
1.164 brouard 3851: int i, j;
1.126 brouard 3852: int *indx;
3853:
3854: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3855: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3856: void lubksb(double **a, int npar, int *indx, double b[]) ;
3857: void ludcmp(double **a, int npar, int *indx, double *d) ;
3858: double gompertz(double p[]);
1.203 brouard 3859: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3860:
3861: printf("\nCalculation of the hessian matrix. Wait...\n");
3862: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3863: for (i=1;i<=npar;i++){
1.203 brouard 3864: printf("%d-",i);fflush(stdout);
3865: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3866:
3867: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3868:
3869: /* printf(" %f ",p[i]);
3870: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3871: }
3872:
3873: for (i=1;i<=npar;i++) {
3874: for (j=1;j<=npar;j++) {
3875: if (j>i) {
1.203 brouard 3876: printf(".%d-%d",i,j);fflush(stdout);
3877: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3878: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3879:
3880: hess[j][i]=hess[i][j];
3881: /*printf(" %lf ",hess[i][j]);*/
3882: }
3883: }
3884: }
3885: printf("\n");
3886: fprintf(ficlog,"\n");
3887:
3888: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3889: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3890:
3891: a=matrix(1,npar,1,npar);
3892: y=matrix(1,npar,1,npar);
3893: x=vector(1,npar);
3894: indx=ivector(1,npar);
3895: for (i=1;i<=npar;i++)
3896: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3897: ludcmp(a,npar,indx,&pd);
3898:
3899: for (j=1;j<=npar;j++) {
3900: for (i=1;i<=npar;i++) x[i]=0;
3901: x[j]=1;
3902: lubksb(a,npar,indx,x);
3903: for (i=1;i<=npar;i++){
3904: matcov[i][j]=x[i];
3905: }
3906: }
3907:
3908: printf("\n#Hessian matrix#\n");
3909: fprintf(ficlog,"\n#Hessian matrix#\n");
3910: for (i=1;i<=npar;i++) {
3911: for (j=1;j<=npar;j++) {
1.203 brouard 3912: printf("%.6e ",hess[i][j]);
3913: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3914: }
3915: printf("\n");
3916: fprintf(ficlog,"\n");
3917: }
3918:
1.203 brouard 3919: /* printf("\n#Covariance matrix#\n"); */
3920: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3921: /* for (i=1;i<=npar;i++) { */
3922: /* for (j=1;j<=npar;j++) { */
3923: /* printf("%.6e ",matcov[i][j]); */
3924: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3925: /* } */
3926: /* printf("\n"); */
3927: /* fprintf(ficlog,"\n"); */
3928: /* } */
3929:
1.126 brouard 3930: /* Recompute Inverse */
1.203 brouard 3931: /* for (i=1;i<=npar;i++) */
3932: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3933: /* ludcmp(a,npar,indx,&pd); */
3934:
3935: /* printf("\n#Hessian matrix recomputed#\n"); */
3936:
3937: /* for (j=1;j<=npar;j++) { */
3938: /* for (i=1;i<=npar;i++) x[i]=0; */
3939: /* x[j]=1; */
3940: /* lubksb(a,npar,indx,x); */
3941: /* for (i=1;i<=npar;i++){ */
3942: /* y[i][j]=x[i]; */
3943: /* printf("%.3e ",y[i][j]); */
3944: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3945: /* } */
3946: /* printf("\n"); */
3947: /* fprintf(ficlog,"\n"); */
3948: /* } */
3949:
3950: /* Verifying the inverse matrix */
3951: #ifdef DEBUGHESS
3952: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3953:
1.203 brouard 3954: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3955: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3956:
3957: for (j=1;j<=npar;j++) {
3958: for (i=1;i<=npar;i++){
1.203 brouard 3959: printf("%.2f ",y[i][j]);
3960: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3961: }
3962: printf("\n");
3963: fprintf(ficlog,"\n");
3964: }
1.203 brouard 3965: #endif
1.126 brouard 3966:
3967: free_matrix(a,1,npar,1,npar);
3968: free_matrix(y,1,npar,1,npar);
3969: free_vector(x,1,npar);
3970: free_ivector(indx,1,npar);
1.203 brouard 3971: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3972:
3973:
3974: }
3975:
3976: /*************** hessian matrix ****************/
3977: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3978: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3979: int i;
3980: int l=1, lmax=20;
1.203 brouard 3981: double k1,k2, res, fx;
1.132 brouard 3982: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3983: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3984: int k=0,kmax=10;
3985: double l1;
3986:
3987: fx=func(x);
3988: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3989: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3990: l1=pow(10,l);
3991: delts=delt;
3992: for(k=1 ; k <kmax; k=k+1){
3993: delt = delta*(l1*k);
3994: p2[theta]=x[theta] +delt;
1.145 brouard 3995: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3996: p2[theta]=x[theta]-delt;
3997: k2=func(p2)-fx;
3998: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3999: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4000:
1.203 brouard 4001: #ifdef DEBUGHESSII
1.126 brouard 4002: 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);
4003: 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);
4004: #endif
4005: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4006: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4007: k=kmax;
4008: }
4009: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4010: k=kmax; l=lmax*10;
1.126 brouard 4011: }
4012: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4013: delts=delt;
4014: }
1.203 brouard 4015: } /* End loop k */
1.126 brouard 4016: }
4017: delti[theta]=delts;
4018: return res;
4019:
4020: }
4021:
1.203 brouard 4022: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4023: {
4024: int i;
1.164 brouard 4025: int l=1, lmax=20;
1.126 brouard 4026: double k1,k2,k3,k4,res,fx;
1.132 brouard 4027: double p2[MAXPARM+1];
1.203 brouard 4028: int k, kmax=1;
4029: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4030:
4031: int firstime=0;
1.203 brouard 4032:
1.126 brouard 4033: fx=func(x);
1.203 brouard 4034: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4035: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4036: p2[thetai]=x[thetai]+delti[thetai]*k;
4037: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4038: k1=func(p2)-fx;
4039:
1.203 brouard 4040: p2[thetai]=x[thetai]+delti[thetai]*k;
4041: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4042: k2=func(p2)-fx;
4043:
1.203 brouard 4044: p2[thetai]=x[thetai]-delti[thetai]*k;
4045: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4046: k3=func(p2)-fx;
4047:
1.203 brouard 4048: p2[thetai]=x[thetai]-delti[thetai]*k;
4049: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4050: k4=func(p2)-fx;
1.203 brouard 4051: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4052: if(k1*k2*k3*k4 <0.){
1.208 brouard 4053: firstime=1;
1.203 brouard 4054: kmax=kmax+10;
1.208 brouard 4055: }
4056: if(kmax >=10 || firstime ==1){
1.246 brouard 4057: 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);
4058: 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 4059: 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);
4060: 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);
4061: }
4062: #ifdef DEBUGHESSIJ
4063: v1=hess[thetai][thetai];
4064: v2=hess[thetaj][thetaj];
4065: cv12=res;
4066: /* Computing eigen value of Hessian matrix */
4067: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4068: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4069: if ((lc2 <0) || (lc1 <0) ){
4070: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4071: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4072: 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);
4073: 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);
4074: }
1.126 brouard 4075: #endif
4076: }
4077: return res;
4078: }
4079:
1.203 brouard 4080: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4081: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4082: /* { */
4083: /* int i; */
4084: /* int l=1, lmax=20; */
4085: /* double k1,k2,k3,k4,res,fx; */
4086: /* double p2[MAXPARM+1]; */
4087: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4088: /* int k=0,kmax=10; */
4089: /* double l1; */
4090:
4091: /* fx=func(x); */
4092: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4093: /* l1=pow(10,l); */
4094: /* delts=delt; */
4095: /* for(k=1 ; k <kmax; k=k+1){ */
4096: /* delt = delti*(l1*k); */
4097: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4098: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4099: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4100: /* k1=func(p2)-fx; */
4101:
4102: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4103: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4104: /* k2=func(p2)-fx; */
4105:
4106: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4107: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4108: /* k3=func(p2)-fx; */
4109:
4110: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4111: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4112: /* k4=func(p2)-fx; */
4113: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4114: /* #ifdef DEBUGHESSIJ */
4115: /* 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); */
4116: /* 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); */
4117: /* #endif */
4118: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4119: /* k=kmax; */
4120: /* } */
4121: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4122: /* k=kmax; l=lmax*10; */
4123: /* } */
4124: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4125: /* delts=delt; */
4126: /* } */
4127: /* } /\* End loop k *\/ */
4128: /* } */
4129: /* delti[theta]=delts; */
4130: /* return res; */
4131: /* } */
4132:
4133:
1.126 brouard 4134: /************** Inverse of matrix **************/
4135: void ludcmp(double **a, int n, int *indx, double *d)
4136: {
4137: int i,imax,j,k;
4138: double big,dum,sum,temp;
4139: double *vv;
4140:
4141: vv=vector(1,n);
4142: *d=1.0;
4143: for (i=1;i<=n;i++) {
4144: big=0.0;
4145: for (j=1;j<=n;j++)
4146: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4147: if (big == 0.0){
4148: printf(" Singular Hessian matrix at row %d:\n",i);
4149: for (j=1;j<=n;j++) {
4150: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4151: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4152: }
4153: fflush(ficlog);
4154: fclose(ficlog);
4155: nrerror("Singular matrix in routine ludcmp");
4156: }
1.126 brouard 4157: vv[i]=1.0/big;
4158: }
4159: for (j=1;j<=n;j++) {
4160: for (i=1;i<j;i++) {
4161: sum=a[i][j];
4162: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4163: a[i][j]=sum;
4164: }
4165: big=0.0;
4166: for (i=j;i<=n;i++) {
4167: sum=a[i][j];
4168: for (k=1;k<j;k++)
4169: sum -= a[i][k]*a[k][j];
4170: a[i][j]=sum;
4171: if ( (dum=vv[i]*fabs(sum)) >= big) {
4172: big=dum;
4173: imax=i;
4174: }
4175: }
4176: if (j != imax) {
4177: for (k=1;k<=n;k++) {
4178: dum=a[imax][k];
4179: a[imax][k]=a[j][k];
4180: a[j][k]=dum;
4181: }
4182: *d = -(*d);
4183: vv[imax]=vv[j];
4184: }
4185: indx[j]=imax;
4186: if (a[j][j] == 0.0) a[j][j]=TINY;
4187: if (j != n) {
4188: dum=1.0/(a[j][j]);
4189: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4190: }
4191: }
4192: free_vector(vv,1,n); /* Doesn't work */
4193: ;
4194: }
4195:
4196: void lubksb(double **a, int n, int *indx, double b[])
4197: {
4198: int i,ii=0,ip,j;
4199: double sum;
4200:
4201: for (i=1;i<=n;i++) {
4202: ip=indx[i];
4203: sum=b[ip];
4204: b[ip]=b[i];
4205: if (ii)
4206: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4207: else if (sum) ii=i;
4208: b[i]=sum;
4209: }
4210: for (i=n;i>=1;i--) {
4211: sum=b[i];
4212: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4213: b[i]=sum/a[i][i];
4214: }
4215: }
4216:
4217: void pstamp(FILE *fichier)
4218: {
1.196 brouard 4219: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4220: }
4221:
1.253 brouard 4222: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4223:
4224: /* y=a+bx regression */
4225: double sumx = 0.0; /* sum of x */
4226: double sumx2 = 0.0; /* sum of x**2 */
4227: double sumxy = 0.0; /* sum of x * y */
4228: double sumy = 0.0; /* sum of y */
4229: double sumy2 = 0.0; /* sum of y**2 */
4230: double sume2; /* sum of square or residuals */
4231: double yhat;
4232:
4233: double denom=0;
4234: int i;
4235: int ne=*no;
4236:
4237: for ( i=ifi, ne=0;i<=ila;i++) {
4238: if(!isfinite(x[i]) || !isfinite(y[i])){
4239: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4240: continue;
4241: }
4242: ne=ne+1;
4243: sumx += x[i];
4244: sumx2 += x[i]*x[i];
4245: sumxy += x[i] * y[i];
4246: sumy += y[i];
4247: sumy2 += y[i]*y[i];
4248: denom = (ne * sumx2 - sumx*sumx);
4249: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4250: }
4251:
4252: denom = (ne * sumx2 - sumx*sumx);
4253: if (denom == 0) {
4254: // vertical, slope m is infinity
4255: *b = INFINITY;
4256: *a = 0;
4257: if (r) *r = 0;
4258: return 1;
4259: }
4260:
4261: *b = (ne * sumxy - sumx * sumy) / denom;
4262: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4263: if (r!=NULL) {
4264: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4265: sqrt((sumx2 - sumx*sumx/ne) *
4266: (sumy2 - sumy*sumy/ne));
4267: }
4268: *no=ne;
4269: for ( i=ifi, ne=0;i<=ila;i++) {
4270: if(!isfinite(x[i]) || !isfinite(y[i])){
4271: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4272: continue;
4273: }
4274: ne=ne+1;
4275: yhat = y[i] - *a -*b* x[i];
4276: sume2 += yhat * yhat ;
4277:
4278: denom = (ne * sumx2 - sumx*sumx);
4279: /* 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); */
4280: }
4281: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4282: *sa= *sb * sqrt(sumx2/ne);
4283:
4284: return 0;
4285: }
4286:
1.126 brouard 4287: /************ Frequencies ********************/
1.251 brouard 4288: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4289: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4290: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4291: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4292:
1.253 brouard 4293: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4294: int iind=0, iage=0;
4295: int mi; /* Effective wave */
4296: int first;
4297: double ***freq; /* Frequencies */
1.253 brouard 4298: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4299: int no;
1.226 brouard 4300: double *meanq;
4301: double **meanqt;
4302: double *pp, **prop, *posprop, *pospropt;
4303: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4304: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4305: double agebegin, ageend;
4306:
4307: pp=vector(1,nlstate);
1.251 brouard 4308: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4309: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4310: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4311: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4312: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4313: meanqt=matrix(1,lastpass,1,nqtveff);
4314: strcpy(fileresp,"P_");
4315: strcat(fileresp,fileresu);
4316: /*strcat(fileresphtm,fileresu);*/
4317: if((ficresp=fopen(fileresp,"w"))==NULL) {
4318: printf("Problem with prevalence resultfile: %s\n", fileresp);
4319: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4320: exit(0);
4321: }
1.240 brouard 4322:
1.226 brouard 4323: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4324: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4325: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4326: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4327: fflush(ficlog);
4328: exit(70);
4329: }
4330: else{
4331: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4332: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4333: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4334: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4335: }
1.237 brouard 4336: 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 4337:
1.226 brouard 4338: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4339: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4340: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4341: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4342: fflush(ficlog);
4343: exit(70);
1.240 brouard 4344: } else{
1.226 brouard 4345: 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 4346: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4347: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4348: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4349: }
1.240 brouard 4350: 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);
4351:
1.253 brouard 4352: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4353: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4354: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4355: j1=0;
1.126 brouard 4356:
1.227 brouard 4357: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4358: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4359: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4360:
4361:
1.226 brouard 4362: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4363: reference=low_education V1=0,V2=0
4364: med_educ V1=1 V2=0,
4365: high_educ V1=0 V2=1
4366: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4367: */
1.249 brouard 4368: dateintsum=0;
4369: k2cpt=0;
4370:
1.253 brouard 4371: if(cptcoveff == 0 )
4372: nl=1; /* Constant model only */
4373: else
4374: nl=2;
4375: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4376: if(nj==1)
4377: j=0; /* First pass for the constant */
4378: else
4379: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4380: first=1;
4381: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on covariates combination in order of model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
4382: posproptt=0.;
4383: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4384: scanf("%d", i);*/
4385: for (i=-5; i<=nlstate+ndeath; i++)
4386: for (jk=-5; jk<=nlstate+ndeath; jk++)
4387: for(m=iagemin; m <= iagemax+3; m++)
4388: freq[i][jk][m]=0;
4389:
4390: for (i=1; i<=nlstate; i++) {
1.240 brouard 4391: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4392: prop[i][m]=0;
4393: posprop[i]=0;
4394: pospropt[i]=0;
4395: }
4396: /* for (z1=1; z1<= nqfveff; z1++) { */
4397: /* meanq[z1]+=0.; */
4398: /* for(m=1;m<=lastpass;m++){ */
4399: /* meanqt[m][z1]=0.; */
4400: /* } */
4401: /* } */
4402:
4403: /* dateintsum=0; */
4404: /* k2cpt=0; */
4405:
4406: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4407: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4408: bool=1;
4409: if(j !=0){
4410: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4411: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4412: /* for (z1=1; z1<= nqfveff; z1++) { */
4413: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4414: /* } */
4415: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4416: /* if(Tvaraff[z1] ==-20){ */
4417: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4418: /* }else if(Tvaraff[z1] ==-10){ */
4419: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4420: /* }else */
4421: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4422: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4423: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4424: /* 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",
4425: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4426: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4427: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4428: } /* Onlyf fixed */
4429: } /* end z1 */
4430: } /* cptcovn > 0 */
4431: } /* end any */
4432: }/* end j==0 */
4433: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4434: /* for(m=firstpass; m<=lastpass; m++){ */
4435: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4436: m=mw[mi][iind];
4437: if(j!=0){
4438: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4439: for (z1=1; z1<=cptcoveff; z1++) {
4440: if( Fixed[Tmodelind[z1]]==1){
4441: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4442: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4443: value is -1, we don't select. It differs from the
4444: constant and age model which counts them. */
4445: bool=0; /* not selected */
4446: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4447: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4448: bool=0;
4449: }
4450: }
4451: }
4452: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4453: } /* end j==0 */
4454: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4455: if(bool==1){
4456: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4457: and mw[mi+1][iind]. dh depends on stepm. */
4458: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4459: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4460: if(m >=firstpass && m <=lastpass){
4461: k2=anint[m][iind]+(mint[m][iind]/12.);
4462: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4463: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4464: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4465: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4466: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4467: if (m<lastpass) {
4468: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4469: /* 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]); */
4470: if(s[m][iind]==-1)
4471: 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.));
4472: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4473: /* if((int)agev[m][iind] == 55) */
4474: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4475: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4476: 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 4477: }
1.251 brouard 4478: } /* end if between passes */
4479: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4480: dateintsum=dateintsum+k2; /* on all covariates ?*/
4481: k2cpt++;
4482: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4483: }
1.251 brouard 4484: }else{
4485: bool=1;
4486: }/* end bool 2 */
4487: } /* end m */
4488: } /* end bool */
4489: } /* end iind = 1 to imx */
4490: /* prop[s][age] is feeded for any initial and valid live state as well as
4491: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4492:
4493:
4494: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4495: pstamp(ficresp);
4496: if (cptcoveff>0 && j!=0){
4497: printf( "\n#********** Variable ");
4498: fprintf(ficresp, "\n#********** Variable ");
4499: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4500: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4501: fprintf(ficlog, "\n#********** Variable ");
4502: for (z1=1; z1<=cptcoveff; z1++){
4503: if(!FixedV[Tvaraff[z1]]){
4504: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4505: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4506: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4507: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4508: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4509: }else{
1.251 brouard 4510: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4511: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4512: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4513: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4514: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4515: }
4516: }
4517: printf( "**********\n#");
4518: fprintf(ficresp, "**********\n#");
4519: fprintf(ficresphtm, "**********</h3>\n");
4520: fprintf(ficresphtmfr, "**********</h3>\n");
4521: fprintf(ficlog, "**********\n");
4522: }
4523: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4524: for(i=1; i<=nlstate;i++) {
4525: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4526: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4527: }
4528: fprintf(ficresp, "\n");
4529: fprintf(ficresphtm, "\n");
4530:
4531: /* Header of frequency table by age */
4532: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4533: fprintf(ficresphtmfr,"<th>Age</th> ");
4534: for(jk=-1; jk <=nlstate+ndeath; jk++){
4535: for(m=-1; m <=nlstate+ndeath; m++){
4536: if(jk!=0 && m!=0)
4537: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4538: }
1.226 brouard 4539: }
1.251 brouard 4540: fprintf(ficresphtmfr, "\n");
4541:
4542: /* For each age */
4543: for(iage=iagemin; iage <= iagemax+3; iage++){
4544: fprintf(ficresphtm,"<tr>");
4545: if(iage==iagemax+1){
4546: fprintf(ficlog,"1");
4547: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4548: }else if(iage==iagemax+2){
4549: fprintf(ficlog,"0");
4550: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4551: }else if(iage==iagemax+3){
4552: fprintf(ficlog,"Total");
4553: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4554: }else{
1.240 brouard 4555: if(first==1){
1.251 brouard 4556: first=0;
4557: printf("See log file for details...\n");
4558: }
4559: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4560: fprintf(ficlog,"Age %d", iage);
4561: }
4562: for(jk=1; jk <=nlstate ; jk++){
4563: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4564: pp[jk] += freq[jk][m][iage];
4565: }
4566: for(jk=1; jk <=nlstate ; jk++){
4567: for(m=-1, pos=0; m <=0 ; m++)
4568: pos += freq[jk][m][iage];
4569: if(pp[jk]>=1.e-10){
4570: if(first==1){
4571: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4572: }
4573: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4574: }else{
4575: if(first==1)
4576: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4577: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4578: }
4579: }
4580:
1.251 brouard 4581: for(jk=1; jk <=nlstate ; jk++){
4582: /* posprop[jk]=0; */
4583: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4584: pp[jk] += freq[jk][m][iage];
4585: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4586:
4587: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4588: pos += pp[jk]; /* pos is the total number of transitions until this age */
4589: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4590: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4591: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4592: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4593: }
1.251 brouard 4594: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4595: if(pos>=1.e-5){
1.251 brouard 4596: if(first==1)
4597: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4598: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4599: }else{
4600: if(first==1)
4601: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4602: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4603: }
4604: if( iage <= iagemax){
4605: if(pos>=1.e-5){
4606: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4607: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4608: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4609: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4610: }
4611: else{
4612: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4613: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4614: }
1.240 brouard 4615: }
1.251 brouard 4616: pospropt[jk] +=posprop[jk];
4617: } /* end loop jk */
4618: /* pospropt=0.; */
4619: for(jk=-1; jk <=nlstate+ndeath; jk++){
4620: for(m=-1; m <=nlstate+ndeath; m++){
4621: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4622: if(first==1){
4623: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4624: }
1.253 brouard 4625: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4626: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4627: }
4628: if(jk!=0 && m!=0)
4629: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4630: }
1.251 brouard 4631: } /* end loop jk */
4632: posproptt=0.;
4633: for(jk=1; jk <=nlstate; jk++){
4634: posproptt += pospropt[jk];
4635: }
4636: fprintf(ficresphtmfr,"</tr>\n ");
4637: if(iage <= iagemax){
4638: fprintf(ficresp,"\n");
4639: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4640: }
1.251 brouard 4641: if(first==1)
4642: printf("Others in log...\n");
4643: fprintf(ficlog,"\n");
4644: } /* end loop age iage */
4645: fprintf(ficresphtm,"<tr><th>Tot</th>");
4646: for(jk=1; jk <=nlstate ; jk++){
4647: if(posproptt < 1.e-5){
4648: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4649: }else{
4650: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4651: }
1.226 brouard 4652: }
1.251 brouard 4653: fprintf(ficresphtm,"</tr>\n");
4654: fprintf(ficresphtm,"</table>\n");
4655: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4656: if(posproptt < 1.e-5){
1.251 brouard 4657: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4658: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4659: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4660: invalidvarcomb[j1]=1;
1.226 brouard 4661: }else{
1.251 brouard 4662: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4663: invalidvarcomb[j1]=0;
1.226 brouard 4664: }
1.251 brouard 4665: fprintf(ficresphtmfr,"</table>\n");
4666: fprintf(ficlog,"\n");
4667: if(j!=0){
4668: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4669: for(i=1,jk=1; i <=nlstate; i++){
4670: for(k=1; k <=(nlstate+ndeath); k++){
4671: if (k != i) {
4672: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4673: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4674: if(j1==1){ /* All dummy covariates to zero */
4675: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4676: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4677: printf("%d%d ",i,k);
4678: fprintf(ficlog,"%d%d ",i,k);
4679: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
4680: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4681: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4682: }
1.253 brouard 4683: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4684: for(iage=iagemin; iage <= iagemax+3; iage++){
4685: x[iage]= (double)iage;
4686: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4687: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4688: }
4689: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4690: pstart[jk]=b;
4691: pstart[jk-1]=a;
1.252 brouard 4692: }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 */
4693: 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]);
4694: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.251 brouard 4695: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4696: printf("%d%d ",i,k);
4697: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4698: printf("jk=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",jk,i,k,jk,p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
4699: }else{ /* Other cases, like quantitative fixed or varying covariates */
4700: ;
4701: }
4702: /* printf("%12.7f )", param[i][jj][k]); */
4703: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4704: jk++;
4705: } /* end jj */
4706: } /* end k!= i */
4707: } /* end k */
4708: } /* end i, jk */
4709: } /* end j !=0 */
4710: } /* end selected combination of covariate j1 */
4711: if(j==0){ /* We can estimate starting values from the occurences in each case */
4712: printf("#Freqsummary: Starting values for the constants:\n");
4713: fprintf(ficlog,"\n");
4714: for(i=1,jk=1; i <=nlstate; i++){
4715: for(k=1; k <=(nlstate+ndeath); k++){
4716: if (k != i) {
4717: printf("%d%d ",i,k);
4718: fprintf(ficlog,"%d%d ",i,k);
4719: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4720: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4721: if(jj==1){ /* Age has to be done */
4722: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4723: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4724: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4725: }
4726: /* printf("%12.7f )", param[i][jj][k]); */
4727: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4728: jk++;
1.250 brouard 4729: }
1.251 brouard 4730: printf("\n");
4731: fprintf(ficlog,"\n");
1.250 brouard 4732: }
4733: }
4734: }
1.251 brouard 4735: printf("#Freqsummary\n");
4736: fprintf(ficlog,"\n");
4737: for(jk=-1; jk <=nlstate+ndeath; jk++){
4738: for(m=-1; m <=nlstate+ndeath; m++){
4739: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4740: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4741: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4742: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4743: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4744: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4745: /* } */
4746: }
4747: } /* end loop jk */
4748:
4749: printf("\n");
4750: fprintf(ficlog,"\n");
4751: } /* end j=0 */
1.249 brouard 4752: } /* end j */
1.252 brouard 4753:
1.253 brouard 4754: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4755: for(i=1, jk=1; i <=nlstate; i++){
4756: for(j=1; j <=nlstate+ndeath; j++){
4757: if(j!=i){
4758: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4759: printf("%1d%1d",i,j);
4760: fprintf(ficparo,"%1d%1d",i,j);
4761: for(k=1; k<=ncovmodel;k++){
4762: /* printf(" %lf",param[i][j][k]); */
4763: /* fprintf(ficparo," %lf",param[i][j][k]); */
4764: p[jk]=pstart[jk];
4765: printf(" %f ",pstart[jk]);
4766: fprintf(ficparo," %f ",pstart[jk]);
4767: jk++;
4768: }
4769: printf("\n");
4770: fprintf(ficparo,"\n");
4771: }
4772: }
4773: }
4774: } /* end mle=-2 */
1.226 brouard 4775: dateintmean=dateintsum/k2cpt;
1.240 brouard 4776:
1.226 brouard 4777: fclose(ficresp);
4778: fclose(ficresphtm);
4779: fclose(ficresphtmfr);
4780: free_vector(meanq,1,nqfveff);
4781: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4782: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4783: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4784: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4785: free_vector(pospropt,1,nlstate);
4786: free_vector(posprop,1,nlstate);
1.251 brouard 4787: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4788: free_vector(pp,1,nlstate);
4789: /* End of freqsummary */
4790: }
1.126 brouard 4791:
4792: /************ Prevalence ********************/
1.227 brouard 4793: 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)
4794: {
4795: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4796: in each health status at the date of interview (if between dateprev1 and dateprev2).
4797: We still use firstpass and lastpass as another selection.
4798: */
1.126 brouard 4799:
1.227 brouard 4800: int i, m, jk, j1, bool, z1,j, iv;
4801: int mi; /* Effective wave */
4802: int iage;
4803: double agebegin, ageend;
4804:
4805: double **prop;
4806: double posprop;
4807: double y2; /* in fractional years */
4808: int iagemin, iagemax;
4809: int first; /** to stop verbosity which is redirected to log file */
4810:
4811: iagemin= (int) agemin;
4812: iagemax= (int) agemax;
4813: /*pp=vector(1,nlstate);*/
1.251 brouard 4814: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4815: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4816: j1=0;
1.222 brouard 4817:
1.227 brouard 4818: /*j=cptcoveff;*/
4819: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4820:
1.227 brouard 4821: first=1;
4822: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4823: for (i=1; i<=nlstate; i++)
1.251 brouard 4824: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4825: prop[i][iage]=0.0;
4826: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4827: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4828: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4829:
4830: for (i=1; i<=imx; i++) { /* Each individual */
4831: bool=1;
4832: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4833: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4834: m=mw[mi][i];
4835: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4836: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4837: for (z1=1; z1<=cptcoveff; z1++){
4838: if( Fixed[Tmodelind[z1]]==1){
4839: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4840: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4841: bool=0;
4842: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4843: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4844: bool=0;
4845: }
4846: }
4847: if(bool==1){ /* Otherwise we skip that wave/person */
4848: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4849: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4850: if(m >=firstpass && m <=lastpass){
4851: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4852: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4853: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4854: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4855: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4856: 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);
4857: exit(1);
4858: }
4859: if (s[m][i]>0 && s[m][i]<=nlstate) {
4860: /*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]]);*/
4861: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4862: prop[s[m][i]][iagemax+3] += weight[i];
4863: } /* end valid statuses */
4864: } /* end selection of dates */
4865: } /* end selection of waves */
4866: } /* end bool */
4867: } /* end wave */
4868: } /* end individual */
4869: for(i=iagemin; i <= iagemax+3; i++){
4870: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4871: posprop += prop[jk][i];
4872: }
4873:
4874: for(jk=1; jk <=nlstate ; jk++){
4875: if( i <= iagemax){
4876: if(posprop>=1.e-5){
4877: probs[i][jk][j1]= prop[jk][i]/posprop;
4878: } else{
4879: if(first==1){
4880: first=0;
4881: printf("Warning Observed prevalence probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,j1,probs[i][jk][j1]);
4882: }
4883: }
4884: }
4885: }/* end jk */
4886: }/* end i */
1.222 brouard 4887: /*} *//* end i1 */
1.227 brouard 4888: } /* end j1 */
1.222 brouard 4889:
1.227 brouard 4890: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4891: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4892: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4893: } /* End of prevalence */
1.126 brouard 4894:
4895: /************* Waves Concatenation ***************/
4896:
4897: 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)
4898: {
4899: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4900: Death is a valid wave (if date is known).
4901: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4902: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4903: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4904: */
1.126 brouard 4905:
1.224 brouard 4906: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4907: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4908: double sum=0., jmean=0.;*/
1.224 brouard 4909: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4910: int j, k=0,jk, ju, jl;
4911: double sum=0.;
4912: first=0;
1.214 brouard 4913: firstwo=0;
1.217 brouard 4914: firsthree=0;
1.218 brouard 4915: firstfour=0;
1.164 brouard 4916: jmin=100000;
1.126 brouard 4917: jmax=-1;
4918: jmean=0.;
1.224 brouard 4919:
4920: /* Treating live states */
1.214 brouard 4921: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4922: mi=0; /* First valid wave */
1.227 brouard 4923: mli=0; /* Last valid wave */
1.126 brouard 4924: m=firstpass;
1.214 brouard 4925: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4926: 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 */
4927: mli=m-1;/* mw[++mi][i]=m-1; */
4928: }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 */
4929: mw[++mi][i]=m;
4930: mli=m;
1.224 brouard 4931: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4932: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4933: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4934: }
1.227 brouard 4935: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4936: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4937: break;
1.224 brouard 4938: #else
1.227 brouard 4939: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4940: if(firsthree == 0){
4941: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as pi. .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m);
4942: firsthree=1;
4943: }
4944: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as pi. .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m);
4945: mw[++mi][i]=m;
4946: mli=m;
4947: }
4948: if(s[m][i]==-2){ /* Vital status is really unknown */
4949: nbwarn++;
4950: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4951: 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);
4952: 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);
4953: }
4954: break;
4955: }
4956: break;
1.224 brouard 4957: #endif
1.227 brouard 4958: }/* End m >= lastpass */
1.126 brouard 4959: }/* end while */
1.224 brouard 4960:
1.227 brouard 4961: /* 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 4962: /* After last pass */
1.224 brouard 4963: /* Treating death states */
1.214 brouard 4964: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4965: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4966: /* } */
1.126 brouard 4967: mi++; /* Death is another wave */
4968: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4969: /* Only death is a correct wave */
1.126 brouard 4970: mw[mi][i]=m;
1.257 brouard 4971: } /* else not in a death state */
1.224 brouard 4972: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 4973: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 4974: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4975: 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 */
4976: nbwarn++;
4977: if(firstfiv==0){
4978: 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 );
4979: firstfiv=1;
4980: }else{
4981: 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 );
4982: }
4983: }else{ /* Death occured afer last wave potential bias */
4984: nberr++;
4985: if(firstwo==0){
1.257 brouard 4986: 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 4987: firstwo=1;
4988: }
1.257 brouard 4989: 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 4990: }
1.257 brouard 4991: }else{ /* if date of interview is unknown */
1.227 brouard 4992: /* death is known but not confirmed by death status at any wave */
4993: if(firstfour==0){
4994: 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 );
4995: firstfour=1;
4996: }
4997: 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 4998: }
1.224 brouard 4999: } /* end if date of death is known */
5000: #endif
5001: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5002: /* wav[i]=mw[mi][i]; */
1.126 brouard 5003: if(mi==0){
5004: nbwarn++;
5005: if(first==0){
1.227 brouard 5006: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5007: first=1;
1.126 brouard 5008: }
5009: if(first==1){
1.227 brouard 5010: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5011: }
5012: } /* end mi==0 */
5013: } /* End individuals */
1.214 brouard 5014: /* wav and mw are no more changed */
1.223 brouard 5015:
1.214 brouard 5016:
1.126 brouard 5017: for(i=1; i<=imx; i++){
5018: for(mi=1; mi<wav[i];mi++){
5019: if (stepm <=0)
1.227 brouard 5020: dh[mi][i]=1;
1.126 brouard 5021: else{
1.227 brouard 5022: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
5023: if (agedc[i] < 2*AGESUP) {
5024: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5025: if(j==0) j=1; /* Survives at least one month after exam */
5026: else if(j<0){
5027: nberr++;
5028: 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]);
5029: j=1; /* Temporary Dangerous patch */
5030: 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);
5031: 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]);
5032: 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);
5033: }
5034: k=k+1;
5035: if (j >= jmax){
5036: jmax=j;
5037: ijmax=i;
5038: }
5039: if (j <= jmin){
5040: jmin=j;
5041: ijmin=i;
5042: }
5043: sum=sum+j;
5044: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5045: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5046: }
5047: }
5048: else{
5049: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5050: /* 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 5051:
1.227 brouard 5052: k=k+1;
5053: if (j >= jmax) {
5054: jmax=j;
5055: ijmax=i;
5056: }
5057: else if (j <= jmin){
5058: jmin=j;
5059: ijmin=i;
5060: }
5061: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5062: /*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]);*/
5063: if(j<0){
5064: nberr++;
5065: 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]);
5066: 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]);
5067: }
5068: sum=sum+j;
5069: }
5070: jk= j/stepm;
5071: jl= j -jk*stepm;
5072: ju= j -(jk+1)*stepm;
5073: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5074: if(jl==0){
5075: dh[mi][i]=jk;
5076: bh[mi][i]=0;
5077: }else{ /* We want a negative bias in order to only have interpolation ie
5078: * to avoid the price of an extra matrix product in likelihood */
5079: dh[mi][i]=jk+1;
5080: bh[mi][i]=ju;
5081: }
5082: }else{
5083: if(jl <= -ju){
5084: dh[mi][i]=jk;
5085: bh[mi][i]=jl; /* bias is positive if real duration
5086: * is higher than the multiple of stepm and negative otherwise.
5087: */
5088: }
5089: else{
5090: dh[mi][i]=jk+1;
5091: bh[mi][i]=ju;
5092: }
5093: if(dh[mi][i]==0){
5094: dh[mi][i]=1; /* At least one step */
5095: bh[mi][i]=ju; /* At least one step */
5096: /* 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);*/
5097: }
5098: } /* end if mle */
1.126 brouard 5099: }
5100: } /* end wave */
5101: }
5102: jmean=sum/k;
5103: 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 5104: 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 5105: }
1.126 brouard 5106:
5107: /*********** Tricode ****************************/
1.220 brouard 5108: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5109: {
5110: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5111: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5112: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5113: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5114: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5115: */
1.130 brouard 5116:
1.242 brouard 5117: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5118: int modmaxcovj=0; /* Modality max of covariates j */
5119: int cptcode=0; /* Modality max of covariates j */
5120: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5121:
5122:
1.242 brouard 5123: /* cptcoveff=0; */
5124: /* *cptcov=0; */
1.126 brouard 5125:
1.242 brouard 5126: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5127:
1.242 brouard 5128: /* Loop on covariates without age and products and no quantitative variable */
5129: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5130: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5131: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5132: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5133: switch(Fixed[k]) {
5134: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5135: 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*/
5136: ij=(int)(covar[Tvar[k]][i]);
5137: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5138: * If product of Vn*Vm, still boolean *:
5139: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5140: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5141: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5142: modality of the nth covariate of individual i. */
5143: if (ij > modmaxcovj)
5144: modmaxcovj=ij;
5145: else if (ij < modmincovj)
5146: modmincovj=ij;
5147: if ((ij < -1) && (ij > NCOVMAX)){
5148: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5149: exit(1);
5150: }else
5151: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5152: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5153: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5154: /* getting the maximum value of the modality of the covariate
5155: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5156: female ies 1, then modmaxcovj=1.
5157: */
5158: } /* end for loop on individuals i */
5159: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5160: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5161: cptcode=modmaxcovj;
5162: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5163: /*for (i=0; i<=cptcode; i++) {*/
5164: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5165: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5166: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5167: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5168: if( j != -1){
5169: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5170: covariate for which somebody answered excluding
5171: undefined. Usually 2: 0 and 1. */
5172: }
5173: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5174: covariate for which somebody answered including
5175: undefined. Usually 3: -1, 0 and 1. */
5176: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5177: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5178: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5179:
1.242 brouard 5180: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5181: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5182: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5183: /* modmincovj=3; modmaxcovj = 7; */
5184: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5185: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5186: /* defining two dummy variables: variables V1_1 and V1_2.*/
5187: /* nbcode[Tvar[j]][ij]=k; */
5188: /* nbcode[Tvar[j]][1]=0; */
5189: /* nbcode[Tvar[j]][2]=1; */
5190: /* nbcode[Tvar[j]][3]=2; */
5191: /* To be continued (not working yet). */
5192: ij=0; /* ij is similar to i but can jump over null modalities */
5193: 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*/
5194: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5195: break;
5196: }
5197: ij++;
5198: 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*/
5199: cptcode = ij; /* New max modality for covar j */
5200: } /* end of loop on modality i=-1 to 1 or more */
5201: break;
5202: case 1: /* Testing on varying covariate, could be simple and
5203: * should look at waves or product of fixed *
5204: * varying. No time to test -1, assuming 0 and 1 only */
5205: ij=0;
5206: for(i=0; i<=1;i++){
5207: nbcode[Tvar[k]][++ij]=i;
5208: }
5209: break;
5210: default:
5211: break;
5212: } /* end switch */
5213: } /* end dummy test */
5214:
5215: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5216: /* /\*recode from 0 *\/ */
5217: /* k is a modality. If we have model=V1+V1*sex */
5218: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5219: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5220: /* } */
5221: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5222: /* if (ij > ncodemax[j]) { */
5223: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5224: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5225: /* break; */
5226: /* } */
5227: /* } /\* end of loop on modality k *\/ */
5228: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5229:
5230: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5231: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5232: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5233: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5234: 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 */
5235: 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 */
5236: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5237: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5238:
5239: ij=0;
5240: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5241: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5242: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5243: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5244: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5245: /* If product not in single variable we don't print results */
5246: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5247: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5248: 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*/
5249: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5250: 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 */
5251: if(Fixed[k]!=0)
5252: anyvaryingduminmodel=1;
5253: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5254: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5255: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5256: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5257: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5258: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5259: }
5260: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5261: /* ij--; */
5262: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5263: *cptcov=ij; /*Number of total real effective covariates: effective
5264: * because they can be excluded from the model and real
5265: * if in the model but excluded because missing values, but how to get k from ij?*/
5266: for(j=ij+1; j<= cptcovt; j++){
5267: Tvaraff[j]=0;
5268: Tmodelind[j]=0;
5269: }
5270: for(j=ntveff+1; j<= cptcovt; j++){
5271: TmodelInvind[j]=0;
5272: }
5273: /* To be sorted */
5274: ;
5275: }
1.126 brouard 5276:
1.145 brouard 5277:
1.126 brouard 5278: /*********** Health Expectancies ****************/
5279:
1.235 brouard 5280: 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 5281:
5282: {
5283: /* Health expectancies, no variances */
1.164 brouard 5284: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5285: int nhstepma, nstepma; /* Decreasing with age */
5286: double age, agelim, hf;
5287: double ***p3mat;
5288: double eip;
5289:
1.238 brouard 5290: /* pstamp(ficreseij); */
1.126 brouard 5291: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5292: fprintf(ficreseij,"# Age");
5293: for(i=1; i<=nlstate;i++){
5294: for(j=1; j<=nlstate;j++){
5295: fprintf(ficreseij," e%1d%1d ",i,j);
5296: }
5297: fprintf(ficreseij," e%1d. ",i);
5298: }
5299: fprintf(ficreseij,"\n");
5300:
5301:
5302: if(estepm < stepm){
5303: printf ("Problem %d lower than %d\n",estepm, stepm);
5304: }
5305: else hstepm=estepm;
5306: /* We compute the life expectancy from trapezoids spaced every estepm months
5307: * This is mainly to measure the difference between two models: for example
5308: * if stepm=24 months pijx are given only every 2 years and by summing them
5309: * we are calculating an estimate of the Life Expectancy assuming a linear
5310: * progression in between and thus overestimating or underestimating according
5311: * to the curvature of the survival function. If, for the same date, we
5312: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5313: * to compare the new estimate of Life expectancy with the same linear
5314: * hypothesis. A more precise result, taking into account a more precise
5315: * curvature will be obtained if estepm is as small as stepm. */
5316:
5317: /* For example we decided to compute the life expectancy with the smallest unit */
5318: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5319: nhstepm is the number of hstepm from age to agelim
5320: nstepm is the number of stepm from age to agelin.
5321: Look at hpijx to understand the reason of that which relies in memory size
5322: and note for a fixed period like estepm months */
5323: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5324: survival function given by stepm (the optimization length). Unfortunately it
5325: means that if the survival funtion is printed only each two years of age and if
5326: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5327: results. So we changed our mind and took the option of the best precision.
5328: */
5329: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5330:
5331: agelim=AGESUP;
5332: /* If stepm=6 months */
5333: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5334: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5335:
5336: /* nhstepm age range expressed in number of stepm */
5337: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5338: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5339: /* if (stepm >= YEARM) hstepm=1;*/
5340: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5341: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5342:
5343: for (age=bage; age<=fage; age ++){
5344: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5345: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5346: /* if (stepm >= YEARM) hstepm=1;*/
5347: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5348:
5349: /* If stepm=6 months */
5350: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5351: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5352:
1.235 brouard 5353: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5354:
5355: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5356:
5357: printf("%d|",(int)age);fflush(stdout);
5358: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5359:
5360: /* Computing expectancies */
5361: for(i=1; i<=nlstate;i++)
5362: for(j=1; j<=nlstate;j++)
5363: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5364: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5365:
5366: /* 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]);*/
5367:
5368: }
5369:
5370: fprintf(ficreseij,"%3.0f",age );
5371: for(i=1; i<=nlstate;i++){
5372: eip=0;
5373: for(j=1; j<=nlstate;j++){
5374: eip +=eij[i][j][(int)age];
5375: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5376: }
5377: fprintf(ficreseij,"%9.4f", eip );
5378: }
5379: fprintf(ficreseij,"\n");
5380:
5381: }
5382: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5383: printf("\n");
5384: fprintf(ficlog,"\n");
5385:
5386: }
5387:
1.235 brouard 5388: 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 5389:
5390: {
5391: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5392: to initial status i, ei. .
1.126 brouard 5393: */
5394: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5395: int nhstepma, nstepma; /* Decreasing with age */
5396: double age, agelim, hf;
5397: double ***p3matp, ***p3matm, ***varhe;
5398: double **dnewm,**doldm;
5399: double *xp, *xm;
5400: double **gp, **gm;
5401: double ***gradg, ***trgradg;
5402: int theta;
5403:
5404: double eip, vip;
5405:
5406: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5407: xp=vector(1,npar);
5408: xm=vector(1,npar);
5409: dnewm=matrix(1,nlstate*nlstate,1,npar);
5410: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5411:
5412: pstamp(ficresstdeij);
5413: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5414: fprintf(ficresstdeij,"# Age");
5415: for(i=1; i<=nlstate;i++){
5416: for(j=1; j<=nlstate;j++)
5417: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5418: fprintf(ficresstdeij," e%1d. ",i);
5419: }
5420: fprintf(ficresstdeij,"\n");
5421:
5422: pstamp(ficrescveij);
5423: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5424: fprintf(ficrescveij,"# Age");
5425: for(i=1; i<=nlstate;i++)
5426: for(j=1; j<=nlstate;j++){
5427: cptj= (j-1)*nlstate+i;
5428: for(i2=1; i2<=nlstate;i2++)
5429: for(j2=1; j2<=nlstate;j2++){
5430: cptj2= (j2-1)*nlstate+i2;
5431: if(cptj2 <= cptj)
5432: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5433: }
5434: }
5435: fprintf(ficrescveij,"\n");
5436:
5437: if(estepm < stepm){
5438: printf ("Problem %d lower than %d\n",estepm, stepm);
5439: }
5440: else hstepm=estepm;
5441: /* We compute the life expectancy from trapezoids spaced every estepm months
5442: * This is mainly to measure the difference between two models: for example
5443: * if stepm=24 months pijx are given only every 2 years and by summing them
5444: * we are calculating an estimate of the Life Expectancy assuming a linear
5445: * progression in between and thus overestimating or underestimating according
5446: * to the curvature of the survival function. If, for the same date, we
5447: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5448: * to compare the new estimate of Life expectancy with the same linear
5449: * hypothesis. A more precise result, taking into account a more precise
5450: * curvature will be obtained if estepm is as small as stepm. */
5451:
5452: /* For example we decided to compute the life expectancy with the smallest unit */
5453: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5454: nhstepm is the number of hstepm from age to agelim
5455: nstepm is the number of stepm from age to agelin.
5456: Look at hpijx to understand the reason of that which relies in memory size
5457: and note for a fixed period like estepm months */
5458: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5459: survival function given by stepm (the optimization length). Unfortunately it
5460: means that if the survival funtion is printed only each two years of age and if
5461: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5462: results. So we changed our mind and took the option of the best precision.
5463: */
5464: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5465:
5466: /* If stepm=6 months */
5467: /* nhstepm age range expressed in number of stepm */
5468: agelim=AGESUP;
5469: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5470: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5471: /* if (stepm >= YEARM) hstepm=1;*/
5472: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5473:
5474: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5475: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5476: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5477: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5478: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5479: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5480:
5481: for (age=bage; age<=fage; age ++){
5482: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5483: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5484: /* if (stepm >= YEARM) hstepm=1;*/
5485: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5486:
1.126 brouard 5487: /* If stepm=6 months */
5488: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5489: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5490:
5491: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5492:
1.126 brouard 5493: /* Computing Variances of health expectancies */
5494: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5495: decrease memory allocation */
5496: for(theta=1; theta <=npar; theta++){
5497: for(i=1; i<=npar; i++){
1.222 brouard 5498: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5499: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5500: }
1.235 brouard 5501: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5502: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5503:
1.126 brouard 5504: for(j=1; j<= nlstate; j++){
1.222 brouard 5505: for(i=1; i<=nlstate; i++){
5506: for(h=0; h<=nhstepm-1; h++){
5507: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5508: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5509: }
5510: }
1.126 brouard 5511: }
1.218 brouard 5512:
1.126 brouard 5513: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5514: for(h=0; h<=nhstepm-1; h++){
5515: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5516: }
1.126 brouard 5517: }/* End theta */
5518:
5519:
5520: for(h=0; h<=nhstepm-1; h++)
5521: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5522: for(theta=1; theta <=npar; theta++)
5523: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5524:
1.218 brouard 5525:
1.222 brouard 5526: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5527: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5528: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5529:
1.222 brouard 5530: printf("%d|",(int)age);fflush(stdout);
5531: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5532: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5533: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5534: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5535: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5536: for(ij=1;ij<=nlstate*nlstate;ij++)
5537: for(ji=1;ji<=nlstate*nlstate;ji++)
5538: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5539: }
5540: }
1.218 brouard 5541:
1.126 brouard 5542: /* Computing expectancies */
1.235 brouard 5543: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5544: for(i=1; i<=nlstate;i++)
5545: for(j=1; j<=nlstate;j++)
1.222 brouard 5546: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5547: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5548:
1.222 brouard 5549: /* 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 5550:
1.222 brouard 5551: }
1.218 brouard 5552:
1.126 brouard 5553: fprintf(ficresstdeij,"%3.0f",age );
5554: for(i=1; i<=nlstate;i++){
5555: eip=0.;
5556: vip=0.;
5557: for(j=1; j<=nlstate;j++){
1.222 brouard 5558: eip += eij[i][j][(int)age];
5559: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5560: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5561: 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 5562: }
5563: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5564: }
5565: fprintf(ficresstdeij,"\n");
1.218 brouard 5566:
1.126 brouard 5567: fprintf(ficrescveij,"%3.0f",age );
5568: for(i=1; i<=nlstate;i++)
5569: for(j=1; j<=nlstate;j++){
1.222 brouard 5570: cptj= (j-1)*nlstate+i;
5571: for(i2=1; i2<=nlstate;i2++)
5572: for(j2=1; j2<=nlstate;j2++){
5573: cptj2= (j2-1)*nlstate+i2;
5574: if(cptj2 <= cptj)
5575: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5576: }
1.126 brouard 5577: }
5578: fprintf(ficrescveij,"\n");
1.218 brouard 5579:
1.126 brouard 5580: }
5581: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5582: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5583: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5584: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5585: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5586: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5587: printf("\n");
5588: fprintf(ficlog,"\n");
1.218 brouard 5589:
1.126 brouard 5590: free_vector(xm,1,npar);
5591: free_vector(xp,1,npar);
5592: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5593: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5594: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5595: }
1.218 brouard 5596:
1.126 brouard 5597: /************ Variance ******************/
1.235 brouard 5598: 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 5599: {
5600: /* Variance of health expectancies */
5601: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5602: /* double **newm;*/
5603: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5604:
5605: /* int movingaverage(); */
5606: double **dnewm,**doldm;
5607: double **dnewmp,**doldmp;
5608: int i, j, nhstepm, hstepm, h, nstepm ;
5609: int k;
5610: double *xp;
5611: double **gp, **gm; /* for var eij */
5612: double ***gradg, ***trgradg; /*for var eij */
5613: double **gradgp, **trgradgp; /* for var p point j */
5614: double *gpp, *gmp; /* for var p point j */
5615: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5616: double ***p3mat;
5617: double age,agelim, hf;
5618: /* double ***mobaverage; */
5619: int theta;
5620: char digit[4];
5621: char digitp[25];
5622:
5623: char fileresprobmorprev[FILENAMELENGTH];
5624:
5625: if(popbased==1){
5626: if(mobilav!=0)
5627: strcpy(digitp,"-POPULBASED-MOBILAV_");
5628: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5629: }
5630: else
5631: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5632:
1.218 brouard 5633: /* if (mobilav!=0) { */
5634: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5635: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5636: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5637: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5638: /* } */
5639: /* } */
5640:
5641: strcpy(fileresprobmorprev,"PRMORPREV-");
5642: sprintf(digit,"%-d",ij);
5643: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5644: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5645: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5646: strcat(fileresprobmorprev,fileresu);
5647: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5648: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5649: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5650: }
5651: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5652: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5653: pstamp(ficresprobmorprev);
5654: 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 5655: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5656: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5657: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5658: }
5659: for(j=1;j<=cptcoveff;j++)
5660: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5661: fprintf(ficresprobmorprev,"\n");
5662:
1.218 brouard 5663: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5664: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5665: fprintf(ficresprobmorprev," p.%-d SE",j);
5666: for(i=1; i<=nlstate;i++)
5667: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5668: }
5669: fprintf(ficresprobmorprev,"\n");
5670:
5671: fprintf(ficgp,"\n# Routine varevsij");
5672: fprintf(ficgp,"\nunset title \n");
5673: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5674: 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");
5675: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5676: /* } */
5677: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5678: pstamp(ficresvij);
5679: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5680: if(popbased==1)
5681: 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);
5682: else
5683: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5684: fprintf(ficresvij,"# Age");
5685: for(i=1; i<=nlstate;i++)
5686: for(j=1; j<=nlstate;j++)
5687: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5688: fprintf(ficresvij,"\n");
5689:
5690: xp=vector(1,npar);
5691: dnewm=matrix(1,nlstate,1,npar);
5692: doldm=matrix(1,nlstate,1,nlstate);
5693: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5694: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5695:
5696: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5697: gpp=vector(nlstate+1,nlstate+ndeath);
5698: gmp=vector(nlstate+1,nlstate+ndeath);
5699: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5700:
1.218 brouard 5701: if(estepm < stepm){
5702: printf ("Problem %d lower than %d\n",estepm, stepm);
5703: }
5704: else hstepm=estepm;
5705: /* For example we decided to compute the life expectancy with the smallest unit */
5706: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5707: nhstepm is the number of hstepm from age to agelim
5708: nstepm is the number of stepm from age to agelim.
5709: Look at function hpijx to understand why because of memory size limitations,
5710: we decided (b) to get a life expectancy respecting the most precise curvature of the
5711: survival function given by stepm (the optimization length). Unfortunately it
5712: means that if the survival funtion is printed every two years of age and if
5713: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5714: results. So we changed our mind and took the option of the best precision.
5715: */
5716: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5717: agelim = AGESUP;
5718: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5719: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5720: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5721: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5722: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5723: gp=matrix(0,nhstepm,1,nlstate);
5724: gm=matrix(0,nhstepm,1,nlstate);
5725:
5726:
5727: for(theta=1; theta <=npar; theta++){
5728: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5729: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5730: }
5731:
1.242 brouard 5732: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5733:
5734: if (popbased==1) {
5735: if(mobilav ==0){
5736: for(i=1; i<=nlstate;i++)
5737: prlim[i][i]=probs[(int)age][i][ij];
5738: }else{ /* mobilav */
5739: for(i=1; i<=nlstate;i++)
5740: prlim[i][i]=mobaverage[(int)age][i][ij];
5741: }
5742: }
5743:
1.235 brouard 5744: 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 5745: for(j=1; j<= nlstate; j++){
5746: for(h=0; h<=nhstepm; h++){
5747: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5748: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5749: }
5750: }
5751: /* Next for computing probability of death (h=1 means
5752: computed over hstepm matrices product = hstepm*stepm months)
5753: as a weighted average of prlim.
5754: */
5755: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5756: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5757: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5758: }
5759: /* end probability of death */
5760:
5761: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5762: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5763:
1.242 brouard 5764: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5765:
5766: if (popbased==1) {
5767: if(mobilav ==0){
5768: for(i=1; i<=nlstate;i++)
5769: prlim[i][i]=probs[(int)age][i][ij];
5770: }else{ /* mobilav */
5771: for(i=1; i<=nlstate;i++)
5772: prlim[i][i]=mobaverage[(int)age][i][ij];
5773: }
5774: }
5775:
1.235 brouard 5776: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5777:
5778: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5779: for(h=0; h<=nhstepm; h++){
5780: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5781: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5782: }
5783: }
5784: /* This for computing probability of death (h=1 means
5785: computed over hstepm matrices product = hstepm*stepm months)
5786: as a weighted average of prlim.
5787: */
5788: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5789: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5790: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5791: }
5792: /* end probability of death */
5793:
5794: for(j=1; j<= nlstate; j++) /* vareij */
5795: for(h=0; h<=nhstepm; h++){
5796: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5797: }
5798:
5799: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5800: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5801: }
5802:
5803: } /* End theta */
5804:
5805: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5806:
5807: for(h=0; h<=nhstepm; h++) /* veij */
5808: for(j=1; j<=nlstate;j++)
5809: for(theta=1; theta <=npar; theta++)
5810: trgradg[h][j][theta]=gradg[h][theta][j];
5811:
5812: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5813: for(theta=1; theta <=npar; theta++)
5814: trgradgp[j][theta]=gradgp[theta][j];
5815:
5816:
5817: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5818: for(i=1;i<=nlstate;i++)
5819: for(j=1;j<=nlstate;j++)
5820: vareij[i][j][(int)age] =0.;
5821:
5822: for(h=0;h<=nhstepm;h++){
5823: for(k=0;k<=nhstepm;k++){
5824: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5825: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5826: for(i=1;i<=nlstate;i++)
5827: for(j=1;j<=nlstate;j++)
5828: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5829: }
5830: }
5831:
5832: /* pptj */
5833: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5834: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5835: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5836: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5837: varppt[j][i]=doldmp[j][i];
5838: /* end ppptj */
5839: /* x centered again */
5840:
1.242 brouard 5841: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5842:
5843: if (popbased==1) {
5844: if(mobilav ==0){
5845: for(i=1; i<=nlstate;i++)
5846: prlim[i][i]=probs[(int)age][i][ij];
5847: }else{ /* mobilav */
5848: for(i=1; i<=nlstate;i++)
5849: prlim[i][i]=mobaverage[(int)age][i][ij];
5850: }
5851: }
5852:
5853: /* This for computing probability of death (h=1 means
5854: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5855: as a weighted average of prlim.
5856: */
1.235 brouard 5857: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5858: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5859: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5860: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5861: }
5862: /* end probability of death */
5863:
5864: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5865: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5866: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5867: for(i=1; i<=nlstate;i++){
5868: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5869: }
5870: }
5871: fprintf(ficresprobmorprev,"\n");
5872:
5873: fprintf(ficresvij,"%.0f ",age );
5874: for(i=1; i<=nlstate;i++)
5875: for(j=1; j<=nlstate;j++){
5876: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5877: }
5878: fprintf(ficresvij,"\n");
5879: free_matrix(gp,0,nhstepm,1,nlstate);
5880: free_matrix(gm,0,nhstepm,1,nlstate);
5881: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5882: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5883: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5884: } /* End age */
5885: free_vector(gpp,nlstate+1,nlstate+ndeath);
5886: free_vector(gmp,nlstate+1,nlstate+ndeath);
5887: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5888: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5889: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5890: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5891: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5892: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5893: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5894: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5895: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5896: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5897: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5898: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5899: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5900: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5901: 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);
5902: /* 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 5903: */
1.218 brouard 5904: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5905: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5906:
1.218 brouard 5907: free_vector(xp,1,npar);
5908: free_matrix(doldm,1,nlstate,1,nlstate);
5909: free_matrix(dnewm,1,nlstate,1,npar);
5910: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5911: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5912: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5913: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5914: fclose(ficresprobmorprev);
5915: fflush(ficgp);
5916: fflush(fichtm);
5917: } /* end varevsij */
1.126 brouard 5918:
5919: /************ Variance of prevlim ******************/
1.235 brouard 5920: 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 5921: {
1.205 brouard 5922: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5923: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5924:
1.126 brouard 5925: double **dnewm,**doldm;
5926: int i, j, nhstepm, hstepm;
5927: double *xp;
5928: double *gp, *gm;
5929: double **gradg, **trgradg;
1.208 brouard 5930: double **mgm, **mgp;
1.126 brouard 5931: double age,agelim;
5932: int theta;
5933:
5934: pstamp(ficresvpl);
5935: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5936: fprintf(ficresvpl,"# Age ");
5937: if(nresult >=1)
5938: fprintf(ficresvpl," Result# ");
1.126 brouard 5939: for(i=1; i<=nlstate;i++)
5940: fprintf(ficresvpl," %1d-%1d",i,i);
5941: fprintf(ficresvpl,"\n");
5942:
5943: xp=vector(1,npar);
5944: dnewm=matrix(1,nlstate,1,npar);
5945: doldm=matrix(1,nlstate,1,nlstate);
5946:
5947: hstepm=1*YEARM; /* Every year of age */
5948: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5949: agelim = AGESUP;
5950: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5951: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5952: if (stepm >= YEARM) hstepm=1;
5953: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5954: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5955: mgp=matrix(1,npar,1,nlstate);
5956: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5957: gp=vector(1,nlstate);
5958: gm=vector(1,nlstate);
5959:
5960: for(theta=1; theta <=npar; theta++){
5961: for(i=1; i<=npar; i++){ /* Computes gradient */
5962: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5963: }
1.209 brouard 5964: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5965: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5966: else
1.235 brouard 5967: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5968: for(i=1;i<=nlstate;i++){
1.126 brouard 5969: gp[i] = prlim[i][i];
1.208 brouard 5970: mgp[theta][i] = prlim[i][i];
5971: }
1.126 brouard 5972: for(i=1; i<=npar; i++) /* Computes gradient */
5973: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5974: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5975: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5976: else
1.235 brouard 5977: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5978: for(i=1;i<=nlstate;i++){
1.126 brouard 5979: gm[i] = prlim[i][i];
1.208 brouard 5980: mgm[theta][i] = prlim[i][i];
5981: }
1.126 brouard 5982: for(i=1;i<=nlstate;i++)
5983: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5984: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5985: } /* End theta */
5986:
5987: trgradg =matrix(1,nlstate,1,npar);
5988:
5989: for(j=1; j<=nlstate;j++)
5990: for(theta=1; theta <=npar; theta++)
5991: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5992: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5993: /* printf("\nmgm mgp %d ",(int)age); */
5994: /* for(j=1; j<=nlstate;j++){ */
5995: /* printf(" %d ",j); */
5996: /* for(theta=1; theta <=npar; theta++) */
5997: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5998: /* printf("\n "); */
5999: /* } */
6000: /* } */
6001: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6002: /* printf("\n gradg %d ",(int)age); */
6003: /* for(j=1; j<=nlstate;j++){ */
6004: /* printf("%d ",j); */
6005: /* for(theta=1; theta <=npar; theta++) */
6006: /* printf("%d %lf ",theta,gradg[theta][j]); */
6007: /* printf("\n "); */
6008: /* } */
6009: /* } */
1.126 brouard 6010:
6011: for(i=1;i<=nlstate;i++)
6012: varpl[i][(int)age] =0.;
1.209 brouard 6013: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6014: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6015: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6016: }else{
1.126 brouard 6017: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6018: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6019: }
1.126 brouard 6020: for(i=1;i<=nlstate;i++)
6021: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6022:
6023: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6024: if(nresult >=1)
6025: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6026: for(i=1; i<=nlstate;i++)
6027: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6028: fprintf(ficresvpl,"\n");
6029: free_vector(gp,1,nlstate);
6030: free_vector(gm,1,nlstate);
1.208 brouard 6031: free_matrix(mgm,1,npar,1,nlstate);
6032: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6033: free_matrix(gradg,1,npar,1,nlstate);
6034: free_matrix(trgradg,1,nlstate,1,npar);
6035: } /* End age */
6036:
6037: free_vector(xp,1,npar);
6038: free_matrix(doldm,1,nlstate,1,npar);
6039: free_matrix(dnewm,1,nlstate,1,nlstate);
6040:
6041: }
6042:
6043: /************ Variance of one-step probabilities ******************/
6044: 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 6045: {
6046: int i, j=0, k1, l1, tj;
6047: int k2, l2, j1, z1;
6048: int k=0, l;
6049: int first=1, first1, first2;
6050: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6051: double **dnewm,**doldm;
6052: double *xp;
6053: double *gp, *gm;
6054: double **gradg, **trgradg;
6055: double **mu;
6056: double age, cov[NCOVMAX+1];
6057: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6058: int theta;
6059: char fileresprob[FILENAMELENGTH];
6060: char fileresprobcov[FILENAMELENGTH];
6061: char fileresprobcor[FILENAMELENGTH];
6062: double ***varpij;
6063:
6064: strcpy(fileresprob,"PROB_");
6065: strcat(fileresprob,fileres);
6066: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6067: printf("Problem with resultfile: %s\n", fileresprob);
6068: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6069: }
6070: strcpy(fileresprobcov,"PROBCOV_");
6071: strcat(fileresprobcov,fileresu);
6072: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6073: printf("Problem with resultfile: %s\n", fileresprobcov);
6074: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6075: }
6076: strcpy(fileresprobcor,"PROBCOR_");
6077: strcat(fileresprobcor,fileresu);
6078: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6079: printf("Problem with resultfile: %s\n", fileresprobcor);
6080: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6081: }
6082: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6083: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6084: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6085: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6086: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6087: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6088: pstamp(ficresprob);
6089: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6090: fprintf(ficresprob,"# Age");
6091: pstamp(ficresprobcov);
6092: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6093: fprintf(ficresprobcov,"# Age");
6094: pstamp(ficresprobcor);
6095: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6096: fprintf(ficresprobcor,"# Age");
1.126 brouard 6097:
6098:
1.222 brouard 6099: for(i=1; i<=nlstate;i++)
6100: for(j=1; j<=(nlstate+ndeath);j++){
6101: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6102: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6103: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6104: }
6105: /* fprintf(ficresprob,"\n");
6106: fprintf(ficresprobcov,"\n");
6107: fprintf(ficresprobcor,"\n");
6108: */
6109: xp=vector(1,npar);
6110: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6111: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6112: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6113: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6114: first=1;
6115: fprintf(ficgp,"\n# Routine varprob");
6116: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6117: fprintf(fichtm,"\n");
6118:
6119: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back.</li>\n",optionfilehtmcov);
6120: 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);
6121: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6122: and drawn. It helps understanding how is the covariance between two incidences.\
6123: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6124: 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 6125: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6126: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6127: standard deviations wide on each axis. <br>\
6128: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6129: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6130: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6131:
1.222 brouard 6132: cov[1]=1;
6133: /* tj=cptcoveff; */
1.225 brouard 6134: tj = (int) pow(2,cptcoveff);
1.222 brouard 6135: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6136: j1=0;
1.224 brouard 6137: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6138: if (cptcovn>0) {
6139: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6140: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6141: fprintf(ficresprob, "**********\n#\n");
6142: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6143: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6144: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6145:
1.222 brouard 6146: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6147: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6148: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6149:
6150:
1.222 brouard 6151: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6152: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6153: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6154:
1.222 brouard 6155: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6156: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6157: fprintf(ficresprobcor, "**********\n#");
6158: if(invalidvarcomb[j1]){
6159: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6160: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6161: continue;
6162: }
6163: }
6164: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6165: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6166: gp=vector(1,(nlstate)*(nlstate+ndeath));
6167: gm=vector(1,(nlstate)*(nlstate+ndeath));
6168: for (age=bage; age<=fage; age ++){
6169: cov[2]=age;
6170: if(nagesqr==1)
6171: cov[3]= age*age;
6172: for (k=1; k<=cptcovn;k++) {
6173: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6174: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6175: * 1 1 1 1 1
6176: * 2 2 1 1 1
6177: * 3 1 2 1 1
6178: */
6179: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6180: }
6181: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6182: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6183: for (k=1; k<=cptcovprod;k++)
6184: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6185:
6186:
1.222 brouard 6187: for(theta=1; theta <=npar; theta++){
6188: for(i=1; i<=npar; i++)
6189: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6190:
1.222 brouard 6191: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6192:
1.222 brouard 6193: k=0;
6194: for(i=1; i<= (nlstate); i++){
6195: for(j=1; j<=(nlstate+ndeath);j++){
6196: k=k+1;
6197: gp[k]=pmmij[i][j];
6198: }
6199: }
1.220 brouard 6200:
1.222 brouard 6201: for(i=1; i<=npar; i++)
6202: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6203:
1.222 brouard 6204: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6205: k=0;
6206: for(i=1; i<=(nlstate); i++){
6207: for(j=1; j<=(nlstate+ndeath);j++){
6208: k=k+1;
6209: gm[k]=pmmij[i][j];
6210: }
6211: }
1.220 brouard 6212:
1.222 brouard 6213: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6214: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6215: }
1.126 brouard 6216:
1.222 brouard 6217: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6218: for(theta=1; theta <=npar; theta++)
6219: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6220:
1.222 brouard 6221: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6222: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6223:
1.222 brouard 6224: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6225:
1.222 brouard 6226: k=0;
6227: for(i=1; i<=(nlstate); i++){
6228: for(j=1; j<=(nlstate+ndeath);j++){
6229: k=k+1;
6230: mu[k][(int) age]=pmmij[i][j];
6231: }
6232: }
6233: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6234: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6235: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6236:
1.222 brouard 6237: /*printf("\n%d ",(int)age);
6238: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6239: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6240: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6241: }*/
1.220 brouard 6242:
1.222 brouard 6243: fprintf(ficresprob,"\n%d ",(int)age);
6244: fprintf(ficresprobcov,"\n%d ",(int)age);
6245: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6246:
1.222 brouard 6247: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6248: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6249: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6250: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6251: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6252: }
6253: i=0;
6254: for (k=1; k<=(nlstate);k++){
6255: for (l=1; l<=(nlstate+ndeath);l++){
6256: i++;
6257: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6258: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6259: for (j=1; j<=i;j++){
6260: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6261: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6262: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6263: }
6264: }
6265: }/* end of loop for state */
6266: } /* end of loop for age */
6267: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6268: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6269: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6270: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6271:
6272: /* Confidence intervalle of pij */
6273: /*
6274: fprintf(ficgp,"\nunset parametric;unset label");
6275: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6276: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6277: 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);
6278: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6279: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6280: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6281: */
6282:
6283: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6284: first1=1;first2=2;
6285: for (k2=1; k2<=(nlstate);k2++){
6286: for (l2=1; l2<=(nlstate+ndeath);l2++){
6287: if(l2==k2) continue;
6288: j=(k2-1)*(nlstate+ndeath)+l2;
6289: for (k1=1; k1<=(nlstate);k1++){
6290: for (l1=1; l1<=(nlstate+ndeath);l1++){
6291: if(l1==k1) continue;
6292: i=(k1-1)*(nlstate+ndeath)+l1;
6293: if(i<=j) continue;
6294: for (age=bage; age<=fage; age ++){
6295: if ((int)age %5==0){
6296: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6297: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6298: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6299: mu1=mu[i][(int) age]/stepm*YEARM ;
6300: mu2=mu[j][(int) age]/stepm*YEARM;
6301: c12=cv12/sqrt(v1*v2);
6302: /* Computing eigen value of matrix of covariance */
6303: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6304: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6305: if ((lc2 <0) || (lc1 <0) ){
6306: if(first2==1){
6307: first1=0;
6308: 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);
6309: }
6310: 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);
6311: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6312: /* lc2=fabs(lc2); */
6313: }
1.220 brouard 6314:
1.222 brouard 6315: /* Eigen vectors */
6316: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6317: /*v21=sqrt(1.-v11*v11); *//* error */
6318: v21=(lc1-v1)/cv12*v11;
6319: v12=-v21;
6320: v22=v11;
6321: tnalp=v21/v11;
6322: if(first1==1){
6323: first1=0;
6324: 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);
6325: }
6326: 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);
6327: /*printf(fignu*/
6328: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6329: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6330: if(first==1){
6331: first=0;
6332: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6333: fprintf(ficgp,"\nset parametric;unset label");
6334: 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);
6335: fprintf(ficgp,"\nset ter svg size 640, 480");
6336: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6337: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6338: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6339: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6340: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6341: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6342: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6343: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6344: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6345: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6346: 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", \
6347: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6348: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6349: }else{
6350: first=0;
6351: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6352: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6353: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6354: 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", \
6355: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6356: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6357: }/* if first */
6358: } /* age mod 5 */
6359: } /* end loop age */
6360: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6361: first=1;
6362: } /*l12 */
6363: } /* k12 */
6364: } /*l1 */
6365: }/* k1 */
6366: } /* loop on combination of covariates j1 */
6367: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6368: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6369: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6370: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6371: free_vector(xp,1,npar);
6372: fclose(ficresprob);
6373: fclose(ficresprobcov);
6374: fclose(ficresprobcor);
6375: fflush(ficgp);
6376: fflush(fichtmcov);
6377: }
1.126 brouard 6378:
6379:
6380: /******************* Printing html file ***********/
1.201 brouard 6381: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6382: int lastpass, int stepm, int weightopt, char model[],\
6383: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6384: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6385: double jprev1, double mprev1,double anprev1, double dateprev1, \
6386: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6387: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6388:
6389: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6390: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6391: </ul>");
1.237 brouard 6392: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6393: </ul>", model);
1.214 brouard 6394: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6395: 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",
6396: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6397: 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 6398: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6399: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6400: fprintf(fichtm,"\
6401: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6402: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6403: fprintf(fichtm,"\
1.217 brouard 6404: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6405: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6406: fprintf(fichtm,"\
1.126 brouard 6407: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6408: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6409: fprintf(fichtm,"\
1.217 brouard 6410: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6411: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6412: fprintf(fichtm,"\
1.211 brouard 6413: - (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 6414: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6415: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6416: if(prevfcast==1){
6417: fprintf(fichtm,"\
6418: - Prevalence projections by age and states: \
1.201 brouard 6419: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6420: }
1.126 brouard 6421:
1.222 brouard 6422: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6423:
1.225 brouard 6424: m=pow(2,cptcoveff);
1.222 brouard 6425: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6426:
1.222 brouard 6427: jj1=0;
1.237 brouard 6428:
6429: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6430: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6431: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6432: continue;
1.220 brouard 6433:
1.222 brouard 6434: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6435: jj1++;
6436: if (cptcovn > 0) {
6437: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6438: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6439: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6440: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6441: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6442: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6443: }
1.237 brouard 6444: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6445: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6446: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6447: }
6448:
1.230 brouard 6449: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6450: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6451: if(invalidvarcomb[k1]){
6452: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6453: printf("\nCombination (%d) ignored because no cases \n",k1);
6454: continue;
6455: }
6456: }
6457: /* aij, bij */
1.259 ! brouard 6458: 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 6459: <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 6460: /* Pij */
1.241 brouard 6461: 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> \
6462: <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 6463: /* Quasi-incidences */
6464: 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 6465: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6466: 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 6467: 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> \
6468: <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 6469: /* Survival functions (period) in state j */
6470: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6471: 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> \
6472: <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 6473: }
6474: /* State specific survival functions (period) */
6475: for(cpt=1; cpt<=nlstate;cpt++){
6476: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6477: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6478: <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 6479: }
6480: /* Period (stable) prevalence in each health state */
6481: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6482: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability to be in state %d some years earlier, knowing that we will be in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6483: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 6484: }
6485: if(backcast==1){
6486: /* Period (stable) back prevalence in each health state */
6487: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6488: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability to be in state %d at a younger age, knowing that we will be in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6489: <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 6490: }
1.217 brouard 6491: }
1.222 brouard 6492: if(prevfcast==1){
6493: /* Projection of prevalence up to period (stable) prevalence in each health state */
6494: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6495: 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> \
6496: <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 6497: }
6498: }
1.220 brouard 6499:
1.222 brouard 6500: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6501: 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> \
6502: <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 6503: }
6504: /* } /\* end i1 *\/ */
6505: }/* End k1 */
6506: fprintf(fichtm,"</ul>");
1.126 brouard 6507:
1.222 brouard 6508: fprintf(fichtm,"\
1.126 brouard 6509: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6510: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6511: - 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 6512: But because parameters are usually highly correlated (a higher incidence of disability \
6513: and a higher incidence of recovery can give very close observed transition) it might \
6514: be very useful to look not only at linear confidence intervals estimated from the \
6515: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6516: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6517: covariance matrix of the one-step probabilities. \
6518: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6519:
1.222 brouard 6520: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6521: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6522: fprintf(fichtm,"\
1.126 brouard 6523: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6524: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6525:
1.222 brouard 6526: fprintf(fichtm,"\
1.126 brouard 6527: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6528: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6529: fprintf(fichtm,"\
1.126 brouard 6530: - 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): \
6531: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6532: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6533: fprintf(fichtm,"\
1.126 brouard 6534: - (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): \
6535: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6536: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6537: fprintf(fichtm,"\
1.128 brouard 6538: - 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 6539: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6540: fprintf(fichtm,"\
1.128 brouard 6541: - 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 6542: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6543: fprintf(fichtm,"\
1.126 brouard 6544: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6545: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6546:
6547: /* if(popforecast==1) fprintf(fichtm,"\n */
6548: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6549: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6550: /* <br>",fileres,fileres,fileres,fileres); */
6551: /* else */
6552: /* 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 6553: fflush(fichtm);
6554: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6555:
1.225 brouard 6556: m=pow(2,cptcoveff);
1.222 brouard 6557: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6558:
1.222 brouard 6559: jj1=0;
1.237 brouard 6560:
1.241 brouard 6561: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6562: for(k1=1; k1<=m;k1++){
1.253 brouard 6563: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6564: continue;
1.222 brouard 6565: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6566: jj1++;
1.126 brouard 6567: if (cptcovn > 0) {
6568: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6569: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6570: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6571: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6572: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6573: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6574: }
6575:
1.126 brouard 6576: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6577:
1.222 brouard 6578: if(invalidvarcomb[k1]){
6579: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6580: continue;
6581: }
1.126 brouard 6582: }
6583: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6584: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6585: 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 6586: <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 6587: }
6588: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6589: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6590: true period expectancies (those weighted with period prevalences are also\
6591: drawn in addition to the population based expectancies computed using\
1.241 brouard 6592: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6593: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6594: /* } /\* end i1 *\/ */
6595: }/* End k1 */
1.241 brouard 6596: }/* End nres */
1.222 brouard 6597: fprintf(fichtm,"</ul>");
6598: fflush(fichtm);
1.126 brouard 6599: }
6600:
6601: /******************* Gnuplot file **************/
1.223 brouard 6602: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6603:
6604: char dirfileres[132],optfileres[132];
1.223 brouard 6605: char gplotcondition[132];
1.237 brouard 6606: 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 6607: int lv=0, vlv=0, kl=0;
1.130 brouard 6608: int ng=0;
1.201 brouard 6609: int vpopbased;
1.223 brouard 6610: int ioffset; /* variable offset for columns */
1.235 brouard 6611: int nres=0; /* Index of resultline */
1.219 brouard 6612:
1.126 brouard 6613: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6614: /* printf("Problem with file %s",optionfilegnuplot); */
6615: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6616: /* } */
6617:
6618: /*#ifdef windows */
6619: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6620: /*#endif */
1.225 brouard 6621: m=pow(2,cptcoveff);
1.126 brouard 6622:
1.202 brouard 6623: /* Contribution to likelihood */
6624: /* Plot the probability implied in the likelihood */
1.223 brouard 6625: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6626: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6627: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6628: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6629: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6630: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6631: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6632: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6633: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6634: 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));
6635: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6636: 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));
6637: for (i=1; i<= nlstate ; i ++) {
6638: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6639: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6640: 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);
6641: for (j=2; j<= nlstate+ndeath ; j ++) {
6642: 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);
6643: }
6644: fprintf(ficgp,";\nset out; unset ylabel;\n");
6645: }
6646: /* 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 */
6647: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6648: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6649: fprintf(ficgp,"\nset out;unset log\n");
6650: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6651:
1.126 brouard 6652: strcpy(dirfileres,optionfilefiname);
6653: strcpy(optfileres,"vpl");
1.223 brouard 6654: /* 1eme*/
1.238 brouard 6655: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6656: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6657: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6658: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6659: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6660: continue;
6661: /* We are interested in selected combination by the resultline */
1.246 brouard 6662: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6663: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6664: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6665: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6666: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6667: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6668: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6669: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6670: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6671: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6672: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6673: }
6674: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6675: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6676: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6677: }
1.246 brouard 6678: /* printf("\n#\n"); */
1.238 brouard 6679: fprintf(ficgp,"\n#\n");
6680: if(invalidvarcomb[k1]){
6681: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6682: continue;
6683: }
1.235 brouard 6684:
1.241 brouard 6685: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6686: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6687: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.235 brouard 6688:
1.238 brouard 6689: for (i=1; i<= nlstate ; i ++) {
6690: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6691: else fprintf(ficgp," %%*lf (%%*lf)");
6692: }
1.242 brouard 6693: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6694: for (i=1; i<= nlstate ; i ++) {
6695: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6696: else fprintf(ficgp," %%*lf (%%*lf)");
6697: }
1.242 brouard 6698: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6699: for (i=1; i<= nlstate ; i ++) {
6700: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6701: else fprintf(ficgp," %%*lf (%%*lf)");
6702: }
6703: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6704: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6705: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6706: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6707: if(cptcoveff ==0){
1.245 brouard 6708: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6709: }else{
6710: kl=0;
6711: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6712: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6713: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6714: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6715: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6716: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6717: kl++;
1.238 brouard 6718: /* 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 *\/ */
6719: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6720: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6721: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6722: if(k==cptcoveff){
1.245 brouard 6723: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6724: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6725: }else{
6726: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6727: kl++;
6728: }
6729: } /* end covariate */
6730: } /* end if no covariate */
6731: } /* end if backcast */
6732: fprintf(ficgp,"\nset out \n");
6733: } /* nres */
1.201 brouard 6734: } /* k1 */
6735: } /* cpt */
1.235 brouard 6736:
6737:
1.126 brouard 6738: /*2 eme*/
1.238 brouard 6739: for (k1=1; k1<= m ; k1 ++){
6740: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6741: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6742: continue;
6743: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6744: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6745: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6746: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6747: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6748: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6749: vlv= nbcode[Tvaraff[k]][lv];
6750: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6751: }
1.237 brouard 6752: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6753: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6754: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6755: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6756: }
1.211 brouard 6757: fprintf(ficgp,"\n#\n");
1.223 brouard 6758: if(invalidvarcomb[k1]){
6759: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6760: continue;
6761: }
1.219 brouard 6762:
1.241 brouard 6763: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6764: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6765: if(vpopbased==0)
6766: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6767: else
6768: fprintf(ficgp,"\nreplot ");
6769: for (i=1; i<= nlstate+1 ; i ++) {
6770: k=2*i;
6771: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1, vpopbased);
6772: for (j=1; j<= nlstate+1 ; j ++) {
6773: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6774: else fprintf(ficgp," %%*lf (%%*lf)");
6775: }
6776: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6777: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6778: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6779: for (j=1; j<= nlstate+1 ; j ++) {
6780: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6781: else fprintf(ficgp," %%*lf (%%*lf)");
6782: }
6783: fprintf(ficgp,"\" t\"\" w l lt 0,");
6784: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6785: for (j=1; j<= nlstate+1 ; j ++) {
6786: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6787: else fprintf(ficgp," %%*lf (%%*lf)");
6788: }
6789: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6790: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6791: } /* state */
6792: } /* vpopbased */
1.244 brouard 6793: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6794: } /* end nres */
6795: } /* k1 end 2 eme*/
6796:
6797:
6798: /*3eme*/
6799: for (k1=1; k1<= m ; k1 ++){
6800: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6801: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6802: continue;
6803:
6804: for (cpt=1; cpt<= nlstate ; cpt ++) {
6805: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6806: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6807: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6808: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6809: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6810: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6811: vlv= nbcode[Tvaraff[k]][lv];
6812: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6813: }
6814: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6815: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6816: }
6817: fprintf(ficgp,"\n#\n");
6818: if(invalidvarcomb[k1]){
6819: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6820: continue;
6821: }
6822:
6823: /* k=2+nlstate*(2*cpt-2); */
6824: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6825: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6826: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6827: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),k1-1,k1-1,k,cpt);
1.238 brouard 6828: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6829: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6830: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6831: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6832: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6833: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6834:
1.238 brouard 6835: */
6836: for (i=1; i< nlstate ; i ++) {
6837: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+i,cpt,i+1);
6838: /* 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 6839:
1.238 brouard 6840: }
6841: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6842: }
6843: } /* end nres */
6844: } /* end kl 3eme */
1.126 brouard 6845:
1.223 brouard 6846: /* 4eme */
1.201 brouard 6847: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6848: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6849: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6850: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6851: continue;
1.238 brouard 6852: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6853: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6854: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6855: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6856: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6857: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6858: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6859: vlv= nbcode[Tvaraff[k]][lv];
6860: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6861: }
6862: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6863: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6864: }
6865: fprintf(ficgp,"\n#\n");
6866: if(invalidvarcomb[k1]){
6867: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6868: continue;
1.223 brouard 6869: }
1.238 brouard 6870:
1.241 brouard 6871: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6872: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6873: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6874: k=3;
6875: for (i=1; i<= nlstate ; i ++){
6876: if(i==1){
6877: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6878: }else{
6879: fprintf(ficgp,", '' ");
6880: }
6881: l=(nlstate+ndeath)*(i-1)+1;
6882: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6883: for (j=2; j<= nlstate+ndeath ; j ++)
6884: fprintf(ficgp,"+$%d",k+l+j-1);
6885: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6886: } /* nlstate */
6887: fprintf(ficgp,"\nset out\n");
6888: } /* end cpt state*/
6889: } /* end nres */
6890: } /* end covariate k1 */
6891:
1.220 brouard 6892: /* 5eme */
1.201 brouard 6893: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6894: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6895: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6896: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6897: continue;
1.238 brouard 6898: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6899: 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);
6900: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6901: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6902: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6903: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6904: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6905: vlv= nbcode[Tvaraff[k]][lv];
6906: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6907: }
6908: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6909: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6910: }
6911: fprintf(ficgp,"\n#\n");
6912: if(invalidvarcomb[k1]){
6913: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6914: continue;
6915: }
1.227 brouard 6916:
1.241 brouard 6917: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6918: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6919: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6920: k=3;
6921: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6922: if(j==1)
6923: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6924: else
6925: fprintf(ficgp,", '' ");
6926: l=(nlstate+ndeath)*(cpt-1) +j;
6927: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6928: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6929: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6930: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6931: } /* nlstate */
6932: fprintf(ficgp,", '' ");
6933: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6934: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6935: l=(nlstate+ndeath)*(cpt-1) +j;
6936: if(j < nlstate)
6937: fprintf(ficgp,"$%d +",k+l);
6938: else
6939: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6940: }
6941: fprintf(ficgp,"\nset out\n");
6942: } /* end cpt state*/
6943: } /* end covariate */
6944: } /* end nres */
1.227 brouard 6945:
1.220 brouard 6946: /* 6eme */
1.202 brouard 6947: /* CV preval stable (period) for each covariate */
1.237 brouard 6948: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6949: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6950: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6951: continue;
1.255 brouard 6952: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6953:
1.211 brouard 6954: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6955: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6956: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6957: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6958: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6959: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6960: vlv= nbcode[Tvaraff[k]][lv];
6961: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6962: }
1.237 brouard 6963: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6964: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6965: }
1.211 brouard 6966: fprintf(ficgp,"\n#\n");
1.223 brouard 6967: if(invalidvarcomb[k1]){
1.227 brouard 6968: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6969: continue;
1.223 brouard 6970: }
1.227 brouard 6971:
1.241 brouard 6972: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6973: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6974: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6975: k=3; /* Offset */
1.255 brouard 6976: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6977: if(i==1)
6978: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6979: else
6980: fprintf(ficgp,", '' ");
1.255 brouard 6981: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6982: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6983: for (j=2; j<= nlstate ; j ++)
6984: fprintf(ficgp,"+$%d",k+l+j-1);
6985: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6986: } /* nlstate */
1.201 brouard 6987: fprintf(ficgp,"\nset out\n");
1.153 brouard 6988: } /* end cpt state*/
6989: } /* end covariate */
1.227 brouard 6990:
6991:
1.220 brouard 6992: /* 7eme */
1.218 brouard 6993: if(backcast == 1){
1.217 brouard 6994: /* CV back preval stable (period) for each covariate */
1.237 brouard 6995: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6996: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6997: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6998: continue;
1.255 brouard 6999: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
7000: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7001: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7002: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7003: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7004: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7005: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7006: vlv= nbcode[Tvaraff[k]][lv];
7007: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7008: }
1.237 brouard 7009: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7010: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7011: }
1.227 brouard 7012: fprintf(ficgp,"\n#\n");
7013: if(invalidvarcomb[k1]){
7014: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7015: continue;
7016: }
7017:
1.241 brouard 7018: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7019: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7020: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7021: k=3; /* Offset */
1.255 brouard 7022: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7023: if(i==1)
7024: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7025: else
7026: fprintf(ficgp,", '' ");
7027: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7028: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7029: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7030: /* 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 7031: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7032: /* for (j=2; j<= nlstate ; j ++) */
7033: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7034: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7035: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7036: } /* nlstate */
7037: fprintf(ficgp,"\nset out\n");
1.218 brouard 7038: } /* end cpt state*/
7039: } /* end covariate */
7040: } /* End if backcast */
7041:
1.223 brouard 7042: /* 8eme */
1.218 brouard 7043: if(prevfcast==1){
7044: /* Projection from cross-sectional to stable (period) for each covariate */
7045:
1.237 brouard 7046: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7047: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7048: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7049: continue;
1.211 brouard 7050: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7051: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7052: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7053: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7054: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7055: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7056: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7057: vlv= nbcode[Tvaraff[k]][lv];
7058: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7059: }
1.237 brouard 7060: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7061: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7062: }
1.227 brouard 7063: fprintf(ficgp,"\n#\n");
7064: if(invalidvarcomb[k1]){
7065: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7066: continue;
7067: }
7068:
7069: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7070: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7071: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7072: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7073: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7074: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7075: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7076: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7077: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7078: if(i==1){
7079: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7080: }else{
7081: fprintf(ficgp,",\\\n '' ");
7082: }
7083: if(cptcoveff ==0){ /* No covariate */
7084: ioffset=2; /* Age is in 2 */
7085: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7086: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7087: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7088: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7089: fprintf(ficgp," u %d:(", ioffset);
7090: if(i==nlstate+1)
7091: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7092: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7093: else
7094: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7095: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7096: }else{ /* more than 2 covariates */
7097: if(cptcoveff ==1){
7098: ioffset=4; /* Age is in 4 */
7099: }else{
7100: ioffset=6; /* Age is in 6 */
7101: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7102: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7103: }
7104: fprintf(ficgp," u %d:(",ioffset);
7105: kl=0;
7106: strcpy(gplotcondition,"(");
7107: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7108: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7109: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7110: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7111: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7112: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7113: kl++;
7114: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7115: kl++;
7116: if(k <cptcoveff && cptcoveff>1)
7117: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7118: }
7119: strcpy(gplotcondition+strlen(gplotcondition),")");
7120: /* 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 *\/ */
7121: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7122: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7123: /* '' 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*/
7124: if(i==nlstate+1){
7125: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7126: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7127: }else{
7128: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7129: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7130: }
7131: } /* end if covariate */
7132: } /* nlstate */
7133: fprintf(ficgp,"\nset out\n");
1.223 brouard 7134: } /* end cpt state*/
7135: } /* end covariate */
7136: } /* End if prevfcast */
1.227 brouard 7137:
7138:
1.238 brouard 7139: /* 9eme writing MLE parameters */
7140: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7141: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7142: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7143: for(k=1; k <=(nlstate+ndeath); k++){
7144: if (k != i) {
1.227 brouard 7145: fprintf(ficgp,"# current state %d\n",k);
7146: for(j=1; j <=ncovmodel; j++){
7147: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7148: jk++;
7149: }
7150: fprintf(ficgp,"\n");
1.126 brouard 7151: }
7152: }
1.223 brouard 7153: }
1.187 brouard 7154: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7155:
1.145 brouard 7156: /*goto avoid;*/
1.238 brouard 7157: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7158: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7159: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7160: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7161: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7162: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7163: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7164: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7165: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7166: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7167: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7168: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7169: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7170: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7171: fprintf(ficgp,"#\n");
1.223 brouard 7172: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7173: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7174: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7175: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7176: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7177: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7178: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7179: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7180: continue;
7181: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7182: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7183: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7184: }
7185: fprintf(ficgp,"\n#\n");
1.241 brouard 7186: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7187: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7188: if (ng==1){
7189: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7190: fprintf(ficgp,"\nunset log y");
7191: }else if (ng==2){
7192: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7193: fprintf(ficgp,"\nset log y");
7194: }else if (ng==3){
7195: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7196: fprintf(ficgp,"\nset log y");
7197: }else
7198: fprintf(ficgp,"\nunset title ");
7199: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7200: i=1;
7201: for(k2=1; k2<=nlstate; k2++) {
7202: k3=i;
7203: for(k=1; k<=(nlstate+ndeath); k++) {
7204: if (k != k2){
7205: switch( ng) {
7206: case 1:
7207: if(nagesqr==0)
7208: fprintf(ficgp," p%d+p%d*x",i,i+1);
7209: else /* nagesqr =1 */
7210: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7211: break;
7212: case 2: /* ng=2 */
7213: if(nagesqr==0)
7214: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7215: else /* nagesqr =1 */
7216: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7217: break;
7218: case 3:
7219: if(nagesqr==0)
7220: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7221: else /* nagesqr =1 */
7222: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7223: break;
7224: }
7225: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7226: ijp=1; /* product no age */
7227: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7228: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7229: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7230: if(j==Tage[ij]) { /* Product by age */
7231: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7232: if(DummyV[j]==0){
1.237 brouard 7233: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7234: }else{ /* quantitative */
7235: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7236: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7237: }
7238: ij++;
7239: }
7240: }else if(j==Tprod[ijp]) { /* */
7241: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7242: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7243: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7244: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7245: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],nbcode[Tvard[ijp][2]][codtabm(jk,j)]); */
7246: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7247: }else{ /* Vn is dummy and Vm is quanti */
7248: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7249: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7250: }
7251: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7252: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7253: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7254: }else{ /* Both quanti */
7255: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7256: }
7257: }
1.238 brouard 7258: ijp++;
1.237 brouard 7259: }
7260: } else{ /* simple covariate */
7261: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7262: if(Dummy[j]==0){
7263: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7264: }else{ /* quantitative */
7265: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7266: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7267: }
1.237 brouard 7268: } /* end simple */
7269: } /* end j */
1.223 brouard 7270: }else{
7271: i=i-ncovmodel;
7272: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7273: fprintf(ficgp," (1.");
7274: }
1.227 brouard 7275:
1.223 brouard 7276: if(ng != 1){
7277: fprintf(ficgp,")/(1");
1.227 brouard 7278:
1.223 brouard 7279: for(k1=1; k1 <=nlstate; k1++){
7280: if(nagesqr==0)
7281: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7282: else /* nagesqr =1 */
7283: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1,k3+(k1-1)*ncovmodel+1+nagesqr);
1.217 brouard 7284:
1.223 brouard 7285: ij=1;
7286: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7287: if((j-2)==Tage[ij]) { /* Bug valgrind */
7288: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7289: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7290: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7291: ij++;
7292: }
7293: }
7294: else
1.225 brouard 7295: fprintf(ficgp,"+p%d*%d",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 7296: }
7297: fprintf(ficgp,")");
7298: }
7299: fprintf(ficgp,")");
7300: if(ng ==2)
7301: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7302: else /* ng= 3 */
7303: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7304: }else{ /* end ng <> 1 */
7305: if( k !=k2) /* logit p11 is hard to draw */
7306: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7307: }
7308: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7309: fprintf(ficgp,",");
7310: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7311: fprintf(ficgp,",");
7312: i=i+ncovmodel;
7313: } /* end k */
7314: } /* end k2 */
7315: fprintf(ficgp,"\n set out\n");
7316: } /* end jk */
7317: } /* end ng */
7318: /* avoid: */
7319: fflush(ficgp);
1.126 brouard 7320: } /* end gnuplot */
7321:
7322:
7323: /*************** Moving average **************/
1.219 brouard 7324: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7325: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7326:
1.222 brouard 7327: int i, cpt, cptcod;
7328: int modcovmax =1;
7329: int mobilavrange, mob;
7330: int iage=0;
7331:
7332: double sum=0.;
7333: double age;
7334: double *sumnewp, *sumnewm;
7335: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7336:
7337:
1.225 brouard 7338: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7339: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7340:
7341: sumnewp = vector(1,ncovcombmax);
7342: sumnewm = vector(1,ncovcombmax);
7343: agemingood = vector(1,ncovcombmax);
7344: agemaxgood = vector(1,ncovcombmax);
7345:
7346: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7347: sumnewm[cptcod]=0.;
7348: sumnewp[cptcod]=0.;
7349: agemingood[cptcod]=0;
7350: agemaxgood[cptcod]=0;
7351: }
7352: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7353:
7354: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7355: if(mobilav==1) mobilavrange=5; /* default */
7356: else mobilavrange=mobilav;
7357: for (age=bage; age<=fage; age++)
7358: for (i=1; i<=nlstate;i++)
7359: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7360: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7361: /* We keep the original values on the extreme ages bage, fage and for
7362: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7363: we use a 5 terms etc. until the borders are no more concerned.
7364: */
7365: for (mob=3;mob <=mobilavrange;mob=mob+2){
7366: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7367: for (i=1; i<=nlstate;i++){
7368: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7369: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7370: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7371: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7372: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7373: }
7374: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7375: }
7376: }
7377: }/* end age */
7378: }/* end mob */
7379: }else
7380: return -1;
7381: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7382: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7383: if(invalidvarcomb[cptcod]){
7384: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7385: continue;
7386: }
1.219 brouard 7387:
1.222 brouard 7388: agemingood[cptcod]=fage-(mob-1)/2;
7389: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7390: sumnewm[cptcod]=0.;
7391: for (i=1; i<=nlstate;i++){
7392: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7393: }
7394: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7395: agemingood[cptcod]=age;
7396: }else{ /* bad */
7397: for (i=1; i<=nlstate;i++){
7398: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7399: } /* i */
7400: } /* end bad */
7401: }/* age */
7402: sum=0.;
7403: for (i=1; i<=nlstate;i++){
7404: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7405: }
7406: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7407: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any descending age!\n",cptcod);
7408: /* for (i=1; i<=nlstate;i++){ */
7409: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7410: /* } /\* i *\/ */
7411: } /* end bad */
7412: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7413: /* From youngest, finding the oldest wrong */
7414: agemaxgood[cptcod]=bage+(mob-1)/2;
7415: for (age=bage+(mob-1)/2; age<=fage; age++){
7416: sumnewm[cptcod]=0.;
7417: for (i=1; i<=nlstate;i++){
7418: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7419: }
7420: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7421: agemaxgood[cptcod]=age;
7422: }else{ /* bad */
7423: for (i=1; i<=nlstate;i++){
7424: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7425: } /* i */
7426: } /* end bad */
7427: }/* age */
7428: sum=0.;
7429: for (i=1; i<=nlstate;i++){
7430: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7431: }
7432: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7433: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any ascending age!\n",cptcod);
7434: /* for (i=1; i<=nlstate;i++){ */
7435: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7436: /* } /\* i *\/ */
7437: } /* end bad */
7438:
7439: for (age=bage; age<=fage; age++){
1.235 brouard 7440: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7441: sumnewp[cptcod]=0.;
7442: sumnewm[cptcod]=0.;
7443: for (i=1; i<=nlstate;i++){
7444: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7445: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7446: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7447: }
7448: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7449: }
7450: /* printf("\n"); */
7451: /* } */
7452: /* brutal averaging */
7453: for (i=1; i<=nlstate;i++){
7454: for (age=1; age<=bage; age++){
7455: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7456: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7457: }
7458: for (age=fage; age<=AGESUP; age++){
7459: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7460: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7461: }
7462: } /* end i status */
7463: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7464: for (age=1; age<=AGESUP; age++){
7465: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7466: mobaverage[(int)age][i][cptcod]=0.;
7467: }
7468: }
7469: }/* end cptcod */
7470: free_vector(sumnewm,1, ncovcombmax);
7471: free_vector(sumnewp,1, ncovcombmax);
7472: free_vector(agemaxgood,1, ncovcombmax);
7473: free_vector(agemingood,1, ncovcombmax);
7474: return 0;
7475: }/* End movingaverage */
1.218 brouard 7476:
1.126 brouard 7477:
7478: /************** Forecasting ******************/
1.235 brouard 7479: 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 7480: /* proj1, year, month, day of starting projection
7481: agemin, agemax range of age
7482: dateprev1 dateprev2 range of dates during which prevalence is computed
7483: anproj2 year of en of projection (same day and month as proj1).
7484: */
1.235 brouard 7485: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7486: double agec; /* generic age */
7487: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7488: double *popeffectif,*popcount;
7489: double ***p3mat;
1.218 brouard 7490: /* double ***mobaverage; */
1.126 brouard 7491: char fileresf[FILENAMELENGTH];
7492:
7493: agelim=AGESUP;
1.211 brouard 7494: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7495: in each health status at the date of interview (if between dateprev1 and dateprev2).
7496: We still use firstpass and lastpass as another selection.
7497: */
1.214 brouard 7498: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7499: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7500:
1.201 brouard 7501: strcpy(fileresf,"F_");
7502: strcat(fileresf,fileresu);
1.126 brouard 7503: if((ficresf=fopen(fileresf,"w"))==NULL) {
7504: printf("Problem with forecast resultfile: %s\n", fileresf);
7505: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7506: }
1.235 brouard 7507: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7508: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7509:
1.225 brouard 7510: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7511:
7512:
7513: stepsize=(int) (stepm+YEARM-1)/YEARM;
7514: if (stepm<=12) stepsize=1;
7515: if(estepm < stepm){
7516: printf ("Problem %d lower than %d\n",estepm, stepm);
7517: }
7518: else hstepm=estepm;
7519:
7520: hstepm=hstepm/stepm;
7521: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7522: fractional in yp1 */
7523: anprojmean=yp;
7524: yp2=modf((yp1*12),&yp);
7525: mprojmean=yp;
7526: yp1=modf((yp2*30.5),&yp);
7527: jprojmean=yp;
7528: if(jprojmean==0) jprojmean=1;
7529: if(mprojmean==0) jprojmean=1;
7530:
1.227 brouard 7531: i1=pow(2,cptcoveff);
1.126 brouard 7532: if (cptcovn < 1){i1=1;}
7533:
7534: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7535:
7536: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7537:
1.126 brouard 7538: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7539: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7540: for(k=1; k<=i1;k++){
1.253 brouard 7541: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7542: continue;
1.227 brouard 7543: if(invalidvarcomb[k]){
7544: printf("\nCombination (%d) projection ignored because no cases \n",k);
7545: continue;
7546: }
7547: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7548: for(j=1;j<=cptcoveff;j++) {
7549: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7550: }
1.235 brouard 7551: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7552: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7553: }
1.227 brouard 7554: fprintf(ficresf," yearproj age");
7555: for(j=1; j<=nlstate+ndeath;j++){
7556: for(i=1; i<=nlstate;i++)
7557: fprintf(ficresf," p%d%d",i,j);
7558: fprintf(ficresf," wp.%d",j);
7559: }
7560: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7561: fprintf(ficresf,"\n");
7562: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7563: for (agec=fage; agec>=(ageminpar-1); agec--){
7564: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7565: nhstepm = nhstepm/hstepm;
7566: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7567: oldm=oldms;savm=savms;
1.235 brouard 7568: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7569:
7570: for (h=0; h<=nhstepm; h++){
7571: if (h*hstepm/YEARM*stepm ==yearp) {
7572: fprintf(ficresf,"\n");
7573: for(j=1;j<=cptcoveff;j++)
7574: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7575: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7576: }
7577: for(j=1; j<=nlstate+ndeath;j++) {
7578: ppij=0.;
7579: for(i=1; i<=nlstate;i++) {
7580: if (mobilav==1)
7581: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7582: else {
7583: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7584: }
7585: if (h*hstepm/YEARM*stepm== yearp) {
7586: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7587: }
7588: } /* end i */
7589: if (h*hstepm/YEARM*stepm==yearp) {
7590: fprintf(ficresf," %.3f", ppij);
7591: }
7592: }/* end j */
7593: } /* end h */
7594: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7595: } /* end agec */
7596: } /* end yearp */
7597: } /* end k */
1.219 brouard 7598:
1.126 brouard 7599: fclose(ficresf);
1.215 brouard 7600: printf("End of Computing forecasting \n");
7601: fprintf(ficlog,"End of Computing forecasting\n");
7602:
1.126 brouard 7603: }
7604:
1.218 brouard 7605: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7606: /* void prevbackforecast(char fileres[], double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){ */
1.218 brouard 7607: /* /\* back1, year, month, day of starting backection */
7608: /* agemin, agemax range of age */
7609: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7610: /* anback2 year of en of backection (same day and month as back1). */
7611: /* *\/ */
7612: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7613: /* double agec; /\* generic age *\/ */
7614: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7615: /* double *popeffectif,*popcount; */
7616: /* double ***p3mat; */
7617: /* /\* double ***mobaverage; *\/ */
7618: /* char fileresfb[FILENAMELENGTH]; */
7619:
7620: /* agelim=AGESUP; */
7621: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7622: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7623: /* We still use firstpass and lastpass as another selection. */
7624: /* *\/ */
7625: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7626: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7627: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7628:
7629: /* strcpy(fileresfb,"FB_"); */
7630: /* strcat(fileresfb,fileresu); */
7631: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7632: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7633: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7634: /* } */
7635: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7636: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7637:
1.225 brouard 7638: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7639:
7640: /* /\* if (mobilav!=0) { *\/ */
7641: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7642: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7643: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7644: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7645: /* /\* } *\/ */
7646: /* /\* } *\/ */
7647:
7648: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7649: /* if (stepm<=12) stepsize=1; */
7650: /* if(estepm < stepm){ */
7651: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7652: /* } */
7653: /* else hstepm=estepm; */
7654:
7655: /* hstepm=hstepm/stepm; */
7656: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7657: /* fractional in yp1 *\/ */
7658: /* anprojmean=yp; */
7659: /* yp2=modf((yp1*12),&yp); */
7660: /* mprojmean=yp; */
7661: /* yp1=modf((yp2*30.5),&yp); */
7662: /* jprojmean=yp; */
7663: /* if(jprojmean==0) jprojmean=1; */
7664: /* if(mprojmean==0) jprojmean=1; */
7665:
1.225 brouard 7666: /* i1=cptcoveff; */
1.218 brouard 7667: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7668:
1.218 brouard 7669: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7670:
1.218 brouard 7671: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7672:
7673: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7674: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7675: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7676: /* k=k+1; */
7677: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7678: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7679: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7680: /* } */
7681: /* fprintf(ficresfb," yearbproj age"); */
7682: /* for(j=1; j<=nlstate+ndeath;j++){ */
7683: /* for(i=1; i<=nlstate;i++) */
7684: /* fprintf(ficresfb," p%d%d",i,j); */
7685: /* fprintf(ficresfb," p.%d",j); */
7686: /* } */
7687: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7688: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7689: /* fprintf(ficresfb,"\n"); */
7690: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7691: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7692: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7693: /* nhstepm = nhstepm/hstepm; */
7694: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7695: /* oldm=oldms;savm=savms; */
7696: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7697: /* for (h=0; h<=nhstepm; h++){ */
7698: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7699: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7700: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7701: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7702: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7703: /* } */
7704: /* for(j=1; j<=nlstate+ndeath;j++) { */
7705: /* ppij=0.; */
7706: /* for(i=1; i<=nlstate;i++) { */
7707: /* if (mobilav==1) */
7708: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7709: /* else { */
7710: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7711: /* } */
7712: /* if (h*hstepm/YEARM*stepm== yearp) { */
7713: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7714: /* } */
7715: /* } /\* end i *\/ */
7716: /* if (h*hstepm/YEARM*stepm==yearp) { */
7717: /* fprintf(ficresfb," %.3f", ppij); */
7718: /* } */
7719: /* }/\* end j *\/ */
7720: /* } /\* end h *\/ */
7721: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7722: /* } /\* end agec *\/ */
7723: /* } /\* end yearp *\/ */
7724: /* } /\* end cptcod *\/ */
7725: /* } /\* end cptcov *\/ */
7726:
7727: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7728:
7729: /* fclose(ficresfb); */
7730: /* printf("End of Computing Back forecasting \n"); */
7731: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7732:
1.218 brouard 7733: /* } */
1.217 brouard 7734:
1.126 brouard 7735: /************** Forecasting *****not tested NB*************/
1.227 brouard 7736: /* 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 7737:
1.227 brouard 7738: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7739: /* int *popage; */
7740: /* double calagedatem, agelim, kk1, kk2; */
7741: /* double *popeffectif,*popcount; */
7742: /* double ***p3mat,***tabpop,***tabpopprev; */
7743: /* /\* double ***mobaverage; *\/ */
7744: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7745:
1.227 brouard 7746: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7747: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7748: /* agelim=AGESUP; */
7749: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7750:
1.227 brouard 7751: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7752:
7753:
1.227 brouard 7754: /* strcpy(filerespop,"POP_"); */
7755: /* strcat(filerespop,fileresu); */
7756: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7757: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7758: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7759: /* } */
7760: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7761: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7762:
1.227 brouard 7763: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7764:
1.227 brouard 7765: /* /\* if (mobilav!=0) { *\/ */
7766: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7767: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7768: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7769: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7770: /* /\* } *\/ */
7771: /* /\* } *\/ */
1.126 brouard 7772:
1.227 brouard 7773: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7774: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7775:
1.227 brouard 7776: /* agelim=AGESUP; */
1.126 brouard 7777:
1.227 brouard 7778: /* hstepm=1; */
7779: /* hstepm=hstepm/stepm; */
1.218 brouard 7780:
1.227 brouard 7781: /* if (popforecast==1) { */
7782: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7783: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7784: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7785: /* } */
7786: /* popage=ivector(0,AGESUP); */
7787: /* popeffectif=vector(0,AGESUP); */
7788: /* popcount=vector(0,AGESUP); */
1.126 brouard 7789:
1.227 brouard 7790: /* i=1; */
7791: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7792:
1.227 brouard 7793: /* imx=i; */
7794: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7795: /* } */
1.218 brouard 7796:
1.227 brouard 7797: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7798: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7799: /* k=k+1; */
7800: /* fprintf(ficrespop,"\n#******"); */
7801: /* for(j=1;j<=cptcoveff;j++) { */
7802: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7803: /* } */
7804: /* fprintf(ficrespop,"******\n"); */
7805: /* fprintf(ficrespop,"# Age"); */
7806: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7807: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7808:
1.227 brouard 7809: /* for (cpt=0; cpt<=0;cpt++) { */
7810: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7811:
1.227 brouard 7812: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7813: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7814: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7815:
1.227 brouard 7816: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7817: /* oldm=oldms;savm=savms; */
7818: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7819:
1.227 brouard 7820: /* for (h=0; h<=nhstepm; h++){ */
7821: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7822: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7823: /* } */
7824: /* for(j=1; j<=nlstate+ndeath;j++) { */
7825: /* kk1=0.;kk2=0; */
7826: /* for(i=1; i<=nlstate;i++) { */
7827: /* if (mobilav==1) */
7828: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7829: /* else { */
7830: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7831: /* } */
7832: /* } */
7833: /* if (h==(int)(calagedatem+12*cpt)){ */
7834: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7835: /* /\*fprintf(ficrespop," %.3f", kk1); */
7836: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7837: /* } */
7838: /* } */
7839: /* for(i=1; i<=nlstate;i++){ */
7840: /* kk1=0.; */
7841: /* for(j=1; j<=nlstate;j++){ */
7842: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7843: /* } */
7844: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7845: /* } */
1.218 brouard 7846:
1.227 brouard 7847: /* if (h==(int)(calagedatem+12*cpt)) */
7848: /* for(j=1; j<=nlstate;j++) */
7849: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7850: /* } */
7851: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7852: /* } */
7853: /* } */
1.218 brouard 7854:
1.227 brouard 7855: /* /\******\/ */
1.218 brouard 7856:
1.227 brouard 7857: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7858: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7859: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7860: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7861: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7862:
1.227 brouard 7863: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7864: /* oldm=oldms;savm=savms; */
7865: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7866: /* for (h=0; h<=nhstepm; h++){ */
7867: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7868: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7869: /* } */
7870: /* for(j=1; j<=nlstate+ndeath;j++) { */
7871: /* kk1=0.;kk2=0; */
7872: /* for(i=1; i<=nlstate;i++) { */
7873: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7874: /* } */
7875: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7876: /* } */
7877: /* } */
7878: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7879: /* } */
7880: /* } */
7881: /* } */
7882: /* } */
1.218 brouard 7883:
1.227 brouard 7884: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7885:
1.227 brouard 7886: /* if (popforecast==1) { */
7887: /* free_ivector(popage,0,AGESUP); */
7888: /* free_vector(popeffectif,0,AGESUP); */
7889: /* free_vector(popcount,0,AGESUP); */
7890: /* } */
7891: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7892: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7893: /* fclose(ficrespop); */
7894: /* } /\* End of popforecast *\/ */
1.218 brouard 7895:
1.126 brouard 7896: int fileappend(FILE *fichier, char *optionfich)
7897: {
7898: if((fichier=fopen(optionfich,"a"))==NULL) {
7899: printf("Problem with file: %s\n", optionfich);
7900: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7901: return (0);
7902: }
7903: fflush(fichier);
7904: return (1);
7905: }
7906:
7907:
7908: /**************** function prwizard **********************/
7909: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7910: {
7911:
7912: /* Wizard to print covariance matrix template */
7913:
1.164 brouard 7914: char ca[32], cb[32];
7915: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7916: int numlinepar;
7917:
7918: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7919: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7920: for(i=1; i <=nlstate; i++){
7921: jj=0;
7922: for(j=1; j <=nlstate+ndeath; j++){
7923: if(j==i) continue;
7924: jj++;
7925: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7926: printf("%1d%1d",i,j);
7927: fprintf(ficparo,"%1d%1d",i,j);
7928: for(k=1; k<=ncovmodel;k++){
7929: /* printf(" %lf",param[i][j][k]); */
7930: /* fprintf(ficparo," %lf",param[i][j][k]); */
7931: printf(" 0.");
7932: fprintf(ficparo," 0.");
7933: }
7934: printf("\n");
7935: fprintf(ficparo,"\n");
7936: }
7937: }
7938: printf("# Scales (for hessian or gradient estimation)\n");
7939: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7940: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7941: for(i=1; i <=nlstate; i++){
7942: jj=0;
7943: for(j=1; j <=nlstate+ndeath; j++){
7944: if(j==i) continue;
7945: jj++;
7946: fprintf(ficparo,"%1d%1d",i,j);
7947: printf("%1d%1d",i,j);
7948: fflush(stdout);
7949: for(k=1; k<=ncovmodel;k++){
7950: /* printf(" %le",delti3[i][j][k]); */
7951: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7952: printf(" 0.");
7953: fprintf(ficparo," 0.");
7954: }
7955: numlinepar++;
7956: printf("\n");
7957: fprintf(ficparo,"\n");
7958: }
7959: }
7960: printf("# Covariance matrix\n");
7961: /* # 121 Var(a12)\n\ */
7962: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7963: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7964: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7965: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7966: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7967: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7968: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7969: fflush(stdout);
7970: fprintf(ficparo,"# Covariance matrix\n");
7971: /* # 121 Var(a12)\n\ */
7972: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7973: /* # ...\n\ */
7974: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7975:
7976: for(itimes=1;itimes<=2;itimes++){
7977: jj=0;
7978: for(i=1; i <=nlstate; i++){
7979: for(j=1; j <=nlstate+ndeath; j++){
7980: if(j==i) continue;
7981: for(k=1; k<=ncovmodel;k++){
7982: jj++;
7983: ca[0]= k+'a'-1;ca[1]='\0';
7984: if(itimes==1){
7985: printf("#%1d%1d%d",i,j,k);
7986: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7987: }else{
7988: printf("%1d%1d%d",i,j,k);
7989: fprintf(ficparo,"%1d%1d%d",i,j,k);
7990: /* printf(" %.5le",matcov[i][j]); */
7991: }
7992: ll=0;
7993: for(li=1;li <=nlstate; li++){
7994: for(lj=1;lj <=nlstate+ndeath; lj++){
7995: if(lj==li) continue;
7996: for(lk=1;lk<=ncovmodel;lk++){
7997: ll++;
7998: if(ll<=jj){
7999: cb[0]= lk +'a'-1;cb[1]='\0';
8000: if(ll<jj){
8001: if(itimes==1){
8002: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8003: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8004: }else{
8005: printf(" 0.");
8006: fprintf(ficparo," 0.");
8007: }
8008: }else{
8009: if(itimes==1){
8010: printf(" Var(%s%1d%1d)",ca,i,j);
8011: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8012: }else{
8013: printf(" 0.");
8014: fprintf(ficparo," 0.");
8015: }
8016: }
8017: }
8018: } /* end lk */
8019: } /* end lj */
8020: } /* end li */
8021: printf("\n");
8022: fprintf(ficparo,"\n");
8023: numlinepar++;
8024: } /* end k*/
8025: } /*end j */
8026: } /* end i */
8027: } /* end itimes */
8028:
8029: } /* end of prwizard */
8030: /******************* Gompertz Likelihood ******************************/
8031: double gompertz(double x[])
8032: {
8033: double A,B,L=0.0,sump=0.,num=0.;
8034: int i,n=0; /* n is the size of the sample */
8035:
1.220 brouard 8036: for (i=1;i<=imx ; i++) {
1.126 brouard 8037: sump=sump+weight[i];
8038: /* sump=sump+1;*/
8039: num=num+1;
8040: }
8041:
8042:
8043: /* for (i=0; i<=imx; i++)
8044: 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]);*/
8045:
8046: for (i=1;i<=imx ; i++)
8047: {
8048: if (cens[i] == 1 && wav[i]>1)
8049: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8050:
8051: if (cens[i] == 0 && wav[i]>1)
8052: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8053: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8054:
8055: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8056: if (wav[i] > 1 ) { /* ??? */
8057: L=L+A*weight[i];
8058: /* 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]);*/
8059: }
8060: }
8061:
8062: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8063:
8064: return -2*L*num/sump;
8065: }
8066:
1.136 brouard 8067: #ifdef GSL
8068: /******************* Gompertz_f Likelihood ******************************/
8069: double gompertz_f(const gsl_vector *v, void *params)
8070: {
8071: double A,B,LL=0.0,sump=0.,num=0.;
8072: double *x= (double *) v->data;
8073: int i,n=0; /* n is the size of the sample */
8074:
8075: for (i=0;i<=imx-1 ; i++) {
8076: sump=sump+weight[i];
8077: /* sump=sump+1;*/
8078: num=num+1;
8079: }
8080:
8081:
8082: /* for (i=0; i<=imx; i++)
8083: 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]);*/
8084: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8085: for (i=1;i<=imx ; i++)
8086: {
8087: if (cens[i] == 1 && wav[i]>1)
8088: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8089:
8090: if (cens[i] == 0 && wav[i]>1)
8091: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8092: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8093:
8094: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8095: if (wav[i] > 1 ) { /* ??? */
8096: LL=LL+A*weight[i];
8097: /* 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]);*/
8098: }
8099: }
8100:
8101: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8102: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8103:
8104: return -2*LL*num/sump;
8105: }
8106: #endif
8107:
1.126 brouard 8108: /******************* Printing html file ***********/
1.201 brouard 8109: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8110: int lastpass, int stepm, int weightopt, char model[],\
8111: int imx, double p[],double **matcov,double agemortsup){
8112: int i,k;
8113:
8114: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8115: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8116: for (i=1;i<=2;i++)
8117: 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 8118: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8119: fprintf(fichtm,"</ul>");
8120:
8121: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8122:
8123: 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>");
8124:
8125: for (k=agegomp;k<(agemortsup-2);k++)
8126: 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]);
8127:
8128:
8129: fflush(fichtm);
8130: }
8131:
8132: /******************* Gnuplot file **************/
1.201 brouard 8133: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8134:
8135: char dirfileres[132],optfileres[132];
1.164 brouard 8136:
1.126 brouard 8137: int ng;
8138:
8139:
8140: /*#ifdef windows */
8141: fprintf(ficgp,"cd \"%s\" \n",pathc);
8142: /*#endif */
8143:
8144:
8145: strcpy(dirfileres,optionfilefiname);
8146: strcpy(optfileres,"vpl");
1.199 brouard 8147: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8148: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8149: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8150: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8151: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8152:
8153: }
8154:
1.136 brouard 8155: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8156: {
1.126 brouard 8157:
1.136 brouard 8158: /*-------- data file ----------*/
8159: FILE *fic;
8160: char dummy[]=" ";
1.240 brouard 8161: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8162: int lstra;
1.136 brouard 8163: int linei, month, year,iout;
8164: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8165: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8166: char *stratrunc;
1.223 brouard 8167:
1.240 brouard 8168: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8169: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8170:
1.240 brouard 8171: for(v=1; v <=ncovcol;v++){
8172: DummyV[v]=0;
8173: FixedV[v]=0;
8174: }
8175: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8176: DummyV[v]=1;
8177: FixedV[v]=0;
8178: }
8179: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8180: DummyV[v]=0;
8181: FixedV[v]=1;
8182: }
8183: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8184: DummyV[v]=1;
8185: FixedV[v]=1;
8186: }
8187: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8188: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8189: 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]);
8190: }
1.126 brouard 8191:
1.136 brouard 8192: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8193: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8194: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8195: }
1.126 brouard 8196:
1.136 brouard 8197: i=1;
8198: linei=0;
8199: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8200: linei=linei+1;
8201: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8202: if(line[j] == '\t')
8203: line[j] = ' ';
8204: }
8205: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8206: ;
8207: };
8208: line[j+1]=0; /* Trims blanks at end of line */
8209: if(line[0]=='#'){
8210: fprintf(ficlog,"Comment line\n%s\n",line);
8211: printf("Comment line\n%s\n",line);
8212: continue;
8213: }
8214: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8215: strcpy(line, linetmp);
1.223 brouard 8216:
8217: /* Loops on waves */
8218: for (j=maxwav;j>=1;j--){
8219: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8220: cutv(stra, strb, line, ' ');
8221: if(strb[0]=='.') { /* Missing value */
8222: lval=-1;
8223: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8224: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8225: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8226: 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);
8227: 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);
8228: return 1;
8229: }
8230: }else{
8231: errno=0;
8232: /* what_kind_of_number(strb); */
8233: dval=strtod(strb,&endptr);
8234: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8235: /* if(strb != endptr && *endptr == '\0') */
8236: /* dval=dlval; */
8237: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8238: if( strb[0]=='\0' || (*endptr != '\0')){
8239: 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);
8240: 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);
8241: return 1;
8242: }
8243: cotqvar[j][iv][i]=dval;
8244: cotvar[j][ntv+iv][i]=dval;
8245: }
8246: strcpy(line,stra);
1.223 brouard 8247: }/* end loop ntqv */
1.225 brouard 8248:
1.223 brouard 8249: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8250: cutv(stra, strb, line, ' ');
8251: if(strb[0]=='.') { /* Missing value */
8252: lval=-1;
8253: }else{
8254: errno=0;
8255: lval=strtol(strb,&endptr,10);
8256: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8257: if( strb[0]=='\0' || (*endptr != '\0')){
8258: 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);
8259: 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);
8260: return 1;
8261: }
8262: }
8263: if(lval <-1 || lval >1){
8264: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8265: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8266: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8267: For example, for multinomial values like 1, 2 and 3,\n \
8268: build V1=0 V2=0 for the reference value (1),\n \
8269: V1=1 V2=0 for (2) \n \
1.223 brouard 8270: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8271: output of IMaCh is often meaningless.\n \
1.223 brouard 8272: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8273: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8274: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8275: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8276: For example, for multinomial values like 1, 2 and 3,\n \
8277: build V1=0 V2=0 for the reference value (1),\n \
8278: V1=1 V2=0 for (2) \n \
1.223 brouard 8279: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8280: output of IMaCh is often meaningless.\n \
1.223 brouard 8281: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8282: return 1;
8283: }
8284: cotvar[j][iv][i]=(double)(lval);
8285: strcpy(line,stra);
1.223 brouard 8286: }/* end loop ntv */
1.225 brouard 8287:
1.223 brouard 8288: /* Statuses at wave */
1.137 brouard 8289: cutv(stra, strb, line, ' ');
1.223 brouard 8290: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8291: lval=-1;
1.136 brouard 8292: }else{
1.238 brouard 8293: errno=0;
8294: lval=strtol(strb,&endptr,10);
8295: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8296: if( strb[0]=='\0' || (*endptr != '\0')){
8297: 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);
8298: 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);
8299: return 1;
8300: }
1.136 brouard 8301: }
1.225 brouard 8302:
1.136 brouard 8303: s[j][i]=lval;
1.225 brouard 8304:
1.223 brouard 8305: /* Date of Interview */
1.136 brouard 8306: strcpy(line,stra);
8307: cutv(stra, strb,line,' ');
1.169 brouard 8308: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8309: }
1.169 brouard 8310: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8311: month=99;
8312: year=9999;
1.136 brouard 8313: }else{
1.225 brouard 8314: 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);
8315: 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);
8316: return 1;
1.136 brouard 8317: }
8318: anint[j][i]= (double) year;
8319: mint[j][i]= (double)month;
8320: strcpy(line,stra);
1.223 brouard 8321: } /* End loop on waves */
1.225 brouard 8322:
1.223 brouard 8323: /* Date of death */
1.136 brouard 8324: cutv(stra, strb,line,' ');
1.169 brouard 8325: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8326: }
1.169 brouard 8327: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8328: month=99;
8329: year=9999;
8330: }else{
1.141 brouard 8331: 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 8332: 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);
8333: return 1;
1.136 brouard 8334: }
8335: andc[i]=(double) year;
8336: moisdc[i]=(double) month;
8337: strcpy(line,stra);
8338:
1.223 brouard 8339: /* Date of birth */
1.136 brouard 8340: cutv(stra, strb,line,' ');
1.169 brouard 8341: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8342: }
1.169 brouard 8343: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8344: month=99;
8345: year=9999;
8346: }else{
1.141 brouard 8347: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);
8348: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);fflush(ficlog);
1.225 brouard 8349: return 1;
1.136 brouard 8350: }
8351: if (year==9999) {
1.141 brouard 8352: 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);
8353: 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 8354: return 1;
8355:
1.136 brouard 8356: }
8357: annais[i]=(double)(year);
8358: moisnais[i]=(double)(month);
8359: strcpy(line,stra);
1.225 brouard 8360:
1.223 brouard 8361: /* Sample weight */
1.136 brouard 8362: cutv(stra, strb,line,' ');
8363: errno=0;
8364: dval=strtod(strb,&endptr);
8365: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8366: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8367: 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 8368: fflush(ficlog);
8369: return 1;
8370: }
8371: weight[i]=dval;
8372: strcpy(line,stra);
1.225 brouard 8373:
1.223 brouard 8374: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8375: cutv(stra, strb, line, ' ');
8376: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8377: lval=-1;
1.223 brouard 8378: }else{
1.225 brouard 8379: errno=0;
8380: /* what_kind_of_number(strb); */
8381: dval=strtod(strb,&endptr);
8382: /* if(strb != endptr && *endptr == '\0') */
8383: /* dval=dlval; */
8384: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8385: if( strb[0]=='\0' || (*endptr != '\0')){
8386: 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);
8387: 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);
8388: return 1;
8389: }
8390: coqvar[iv][i]=dval;
1.226 brouard 8391: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8392: }
8393: strcpy(line,stra);
8394: }/* end loop nqv */
1.136 brouard 8395:
1.223 brouard 8396: /* Covariate values */
1.136 brouard 8397: for (j=ncovcol;j>=1;j--){
8398: cutv(stra, strb,line,' ');
1.223 brouard 8399: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8400: lval=-1;
1.136 brouard 8401: }else{
1.225 brouard 8402: errno=0;
8403: lval=strtol(strb,&endptr,10);
8404: if( strb[0]=='\0' || (*endptr != '\0')){
8405: 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);
8406: 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);
8407: return 1;
8408: }
1.136 brouard 8409: }
8410: if(lval <-1 || lval >1){
1.225 brouard 8411: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8412: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8413: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8414: For example, for multinomial values like 1, 2 and 3,\n \
8415: build V1=0 V2=0 for the reference value (1),\n \
8416: V1=1 V2=0 for (2) \n \
1.136 brouard 8417: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8418: output of IMaCh is often meaningless.\n \
1.136 brouard 8419: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8420: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8421: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8422: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8423: For example, for multinomial values like 1, 2 and 3,\n \
8424: build V1=0 V2=0 for the reference value (1),\n \
8425: V1=1 V2=0 for (2) \n \
1.136 brouard 8426: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8427: output of IMaCh is often meaningless.\n \
1.136 brouard 8428: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8429: return 1;
1.136 brouard 8430: }
8431: covar[j][i]=(double)(lval);
8432: strcpy(line,stra);
8433: }
8434: lstra=strlen(stra);
1.225 brouard 8435:
1.136 brouard 8436: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8437: stratrunc = &(stra[lstra-9]);
8438: num[i]=atol(stratrunc);
8439: }
8440: else
8441: num[i]=atol(stra);
8442: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8443: 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;}*/
8444:
8445: i=i+1;
8446: } /* End loop reading data */
1.225 brouard 8447:
1.136 brouard 8448: *imax=i-1; /* Number of individuals */
8449: fclose(fic);
1.225 brouard 8450:
1.136 brouard 8451: return (0);
1.164 brouard 8452: /* endread: */
1.225 brouard 8453: printf("Exiting readdata: ");
8454: fclose(fic);
8455: return (1);
1.223 brouard 8456: }
1.126 brouard 8457:
1.234 brouard 8458: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8459: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8460: while (*p2 == ' ')
1.234 brouard 8461: p2++;
8462: /* while ((*p1++ = *p2++) !=0) */
8463: /* ; */
8464: /* do */
8465: /* while (*p2 == ' ') */
8466: /* p2++; */
8467: /* while (*p1++ == *p2++); */
8468: *stri=p2;
1.145 brouard 8469: }
8470:
1.235 brouard 8471: int decoderesult ( char resultline[], int nres)
1.230 brouard 8472: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8473: {
1.235 brouard 8474: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8475: char resultsav[MAXLINE];
1.234 brouard 8476: int resultmodel[MAXLINE];
8477: int modelresult[MAXLINE];
1.230 brouard 8478: char stra[80], strb[80], strc[80], strd[80],stre[80];
8479:
1.234 brouard 8480: removefirstspace(&resultline);
1.233 brouard 8481: printf("decoderesult:%s\n",resultline);
1.230 brouard 8482:
8483: if (strstr(resultline,"v") !=0){
8484: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8485: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8486: return 1;
8487: }
8488: trimbb(resultsav, resultline);
8489: if (strlen(resultsav) >1){
8490: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8491: }
1.253 brouard 8492: if(j == 0){ /* Resultline but no = */
8493: TKresult[nres]=0; /* Combination for the nresult and the model */
8494: return (0);
8495: }
8496:
1.234 brouard 8497: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8498: 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);
8499: 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);
8500: }
8501: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8502: if(nbocc(resultsav,'=') >1){
8503: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8504: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8505: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8506: }else
8507: cutl(strc,strd,resultsav,'=');
1.230 brouard 8508: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8509:
1.230 brouard 8510: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8511: Tvarsel[k]=atoi(strc);
8512: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8513: /* cptcovsel++; */
8514: if (nbocc(stra,'=') >0)
8515: strcpy(resultsav,stra); /* and analyzes it */
8516: }
1.235 brouard 8517: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8518: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8519: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8520: match=0;
1.236 brouard 8521: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8522: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8523: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8524: match=1;
8525: break;
8526: }
8527: }
8528: if(match == 0){
8529: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8530: }
8531: }
8532: }
1.235 brouard 8533: /* Checking for missing or useless values in comparison of current model needs */
8534: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8535: match=0;
1.235 brouard 8536: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8537: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8538: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8539: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8540: ++match;
8541: }
8542: }
8543: }
8544: if(match == 0){
8545: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8546: }else if(match > 1){
8547: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8548: }
8549: }
1.235 brouard 8550:
1.234 brouard 8551: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8552: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8553: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8554: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8555: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8556: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8557: /* 1 0 0 0 */
8558: /* 2 1 0 0 */
8559: /* 3 0 1 0 */
8560: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8561: /* 5 0 0 1 */
8562: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8563: /* 7 0 1 1 */
8564: /* 8 1 1 1 */
1.237 brouard 8565: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8566: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8567: /* V5*age V5 known which value for nres? */
8568: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8569: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8570: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8571: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8572: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8573: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8574: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8575: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8576: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8577: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8578: k4++;;
8579: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8580: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8581: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8582: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8583: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8584: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8585: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8586: k4q++;;
8587: }
8588: }
1.234 brouard 8589:
1.235 brouard 8590: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8591: return (0);
8592: }
1.235 brouard 8593:
1.230 brouard 8594: int decodemodel( char model[], int lastobs)
8595: /**< This routine decodes the model and returns:
1.224 brouard 8596: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8597: * - nagesqr = 1 if age*age in the model, otherwise 0.
8598: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8599: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8600: * - cptcovage number of covariates with age*products =2
8601: * - cptcovs number of simple covariates
8602: * - 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
8603: * which is a new column after the 9 (ncovcol) variables.
8604: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8605: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8606: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8607: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8608: */
1.136 brouard 8609: {
1.238 brouard 8610: int i, j, k, ks, v;
1.227 brouard 8611: int j1, k1, k2, k3, k4;
1.136 brouard 8612: char modelsav[80];
1.145 brouard 8613: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8614: char *strpt;
1.136 brouard 8615:
1.145 brouard 8616: /*removespace(model);*/
1.136 brouard 8617: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8618: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8619: if (strstr(model,"AGE") !=0){
1.192 brouard 8620: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8621: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8622: return 1;
8623: }
1.141 brouard 8624: if (strstr(model,"v") !=0){
8625: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8626: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8627: return 1;
8628: }
1.187 brouard 8629: strcpy(modelsav,model);
8630: if ((strpt=strstr(model,"age*age")) !=0){
8631: printf(" strpt=%s, model=%s\n",strpt, model);
8632: if(strpt != model){
1.234 brouard 8633: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8634: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8635: corresponding column of parameters.\n",model);
1.234 brouard 8636: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8637: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8638: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8639: return 1;
1.225 brouard 8640: }
1.187 brouard 8641: nagesqr=1;
8642: if (strstr(model,"+age*age") !=0)
1.234 brouard 8643: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8644: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8645: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8646: else
1.234 brouard 8647: substrchaine(modelsav, model, "age*age");
1.187 brouard 8648: }else
8649: nagesqr=0;
8650: if (strlen(modelsav) >1){
8651: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8652: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8653: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8654: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8655: * cst, age and age*age
8656: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8657: /* including age products which are counted in cptcovage.
8658: * but the covariates which are products must be treated
8659: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8660: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8661: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8662:
8663:
1.187 brouard 8664: /* Design
8665: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8666: * < ncovcol=8 >
8667: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8668: * k= 1 2 3 4 5 6 7 8
8669: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8670: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8671: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8672: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8673: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8674: * Tage[++cptcovage]=k
8675: * if products, new covar are created after ncovcol with k1
8676: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8677: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8678: * 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
8679: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8680: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8681: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8682: * < ncovcol=8 >
8683: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8684: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8685: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8686: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8687: * p Tprod[1]@2={ 6, 5}
8688: *p Tvard[1][1]@4= {7, 8, 5, 6}
8689: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8690: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8691: *How to reorganize?
8692: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8693: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8694: * {2, 1, 4, 8, 5, 6, 3, 7}
8695: * Struct []
8696: */
1.225 brouard 8697:
1.187 brouard 8698: /* This loop fills the array Tvar from the string 'model'.*/
8699: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8700: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8701: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8702: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8703: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8704: /* k=1 Tvar[1]=2 (from V2) */
8705: /* k=5 Tvar[5] */
8706: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8707: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8708: /* } */
1.198 brouard 8709: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8710: /*
8711: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8712: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8713: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8714: }
1.187 brouard 8715: cptcovage=0;
8716: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8717: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8718: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8719: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8720: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8721: /*scanf("%d",i);*/
8722: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8723: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8724: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8725: /* covar is not filled and then is empty */
8726: cptcovprod--;
8727: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8728: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8729: Typevar[k]=1; /* 1 for age product */
8730: cptcovage++; /* Sums the number of covariates which include age as a product */
8731: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8732: /*printf("stre=%s ", stre);*/
8733: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8734: cptcovprod--;
8735: cutl(stre,strb,strc,'V');
8736: Tvar[k]=atoi(stre);
8737: Typevar[k]=1; /* 1 for age product */
8738: cptcovage++;
8739: Tage[cptcovage]=k;
8740: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8741: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8742: cptcovn++;
8743: cptcovprodnoage++;k1++;
8744: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8745: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8746: because this model-covariate is a construction we invent a new column
8747: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8748: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8749: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8750: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8751: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8752: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8753: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8754: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8755: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8756: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8757: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8758: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8759: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8760: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8761: for (i=1; i<=lastobs;i++){
8762: /* Computes the new covariate which is a product of
8763: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8764: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8765: }
8766: } /* End age is not in the model */
8767: } /* End if model includes a product */
8768: else { /* no more sum */
8769: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8770: /* scanf("%d",i);*/
8771: cutl(strd,strc,strb,'V');
8772: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8773: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8774: Tvar[k]=atoi(strd);
8775: Typevar[k]=0; /* 0 for simple covariates */
8776: }
8777: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8778: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8779: scanf("%d",i);*/
1.187 brouard 8780: } /* end of loop + on total covariates */
8781: } /* end if strlen(modelsave == 0) age*age might exist */
8782: } /* end if strlen(model == 0) */
1.136 brouard 8783:
8784: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8785: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8786:
1.136 brouard 8787: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8788: printf("cptcovprod=%d ", cptcovprod);
8789: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8790: scanf("%d ",i);*/
8791:
8792:
1.230 brouard 8793: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8794: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8795: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8796: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8797: k = 1 2 3 4 5 6 7 8 9
8798: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8799: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8800: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8801: Dummy[k] 1 0 0 0 3 1 1 2 3
8802: Tmodelind[combination of covar]=k;
1.225 brouard 8803: */
8804: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8805: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8806: /* 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 8807: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8808: printf("Model=%s\n\
8809: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8810: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8811: 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);
8812: fprintf(ficlog,"Model=%s\n\
8813: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8814: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8815: 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 8816: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8817: 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 */
8818: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8819: Fixed[k]= 0;
8820: Dummy[k]= 0;
1.225 brouard 8821: ncoveff++;
1.232 brouard 8822: ncovf++;
1.234 brouard 8823: nsd++;
8824: modell[k].maintype= FTYPE;
8825: TvarsD[nsd]=Tvar[k];
8826: TvarsDind[nsd]=k;
8827: TvarF[ncovf]=Tvar[k];
8828: TvarFind[ncovf]=k;
8829: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8830: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8831: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8832: Fixed[k]= 0;
8833: Dummy[k]= 0;
8834: ncoveff++;
8835: ncovf++;
8836: modell[k].maintype= FTYPE;
8837: TvarF[ncovf]=Tvar[k];
8838: TvarFind[ncovf]=k;
1.230 brouard 8839: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8840: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8841: }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 8842: Fixed[k]= 0;
8843: Dummy[k]= 1;
1.230 brouard 8844: nqfveff++;
1.234 brouard 8845: modell[k].maintype= FTYPE;
8846: modell[k].subtype= FQ;
8847: nsq++;
8848: TvarsQ[nsq]=Tvar[k];
8849: TvarsQind[nsq]=k;
1.232 brouard 8850: ncovf++;
1.234 brouard 8851: TvarF[ncovf]=Tvar[k];
8852: TvarFind[ncovf]=k;
1.231 brouard 8853: 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 8854: 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 8855: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8856: Fixed[k]= 1;
8857: Dummy[k]= 0;
1.225 brouard 8858: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8859: modell[k].maintype= VTYPE;
8860: modell[k].subtype= VD;
8861: nsd++;
8862: TvarsD[nsd]=Tvar[k];
8863: TvarsDind[nsd]=k;
8864: ncovv++; /* Only simple time varying variables */
8865: TvarV[ncovv]=Tvar[k];
1.242 brouard 8866: 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 8867: 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 */
8868: 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 8869: 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);
8870: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8871: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8872: Fixed[k]= 1;
8873: Dummy[k]= 1;
8874: nqtveff++;
8875: modell[k].maintype= VTYPE;
8876: modell[k].subtype= VQ;
8877: ncovv++; /* Only simple time varying variables */
8878: nsq++;
8879: TvarsQ[nsq]=Tvar[k];
8880: TvarsQind[nsq]=k;
8881: TvarV[ncovv]=Tvar[k];
1.242 brouard 8882: 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 8883: 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 */
8884: 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 8885: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8886: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8887: 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 8888: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8889: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8890: ncova++;
8891: TvarA[ncova]=Tvar[k];
8892: TvarAind[ncova]=k;
1.231 brouard 8893: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8894: Fixed[k]= 2;
8895: Dummy[k]= 2;
8896: modell[k].maintype= ATYPE;
8897: modell[k].subtype= APFD;
8898: /* ncoveff++; */
1.227 brouard 8899: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8900: Fixed[k]= 2;
8901: Dummy[k]= 3;
8902: modell[k].maintype= ATYPE;
8903: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8904: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8905: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8906: Fixed[k]= 3;
8907: Dummy[k]= 2;
8908: modell[k].maintype= ATYPE;
8909: modell[k].subtype= APVD; /* Product age * varying dummy */
8910: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8911: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8912: Fixed[k]= 3;
8913: Dummy[k]= 3;
8914: modell[k].maintype= ATYPE;
8915: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8916: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8917: }
8918: }else if (Typevar[k] == 2) { /* product without age */
8919: k1=Tposprod[k];
8920: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8921: if(Tvard[k1][2] <=ncovcol){
8922: Fixed[k]= 1;
8923: Dummy[k]= 0;
8924: modell[k].maintype= FTYPE;
8925: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8926: ncovf++; /* Fixed variables without age */
8927: TvarF[ncovf]=Tvar[k];
8928: TvarFind[ncovf]=k;
8929: }else if(Tvard[k1][2] <=ncovcol+nqv){
8930: Fixed[k]= 0; /* or 2 ?*/
8931: Dummy[k]= 1;
8932: modell[k].maintype= FTYPE;
8933: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8934: ncovf++; /* Varying variables without age */
8935: TvarF[ncovf]=Tvar[k];
8936: TvarFind[ncovf]=k;
8937: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8938: Fixed[k]= 1;
8939: Dummy[k]= 0;
8940: modell[k].maintype= VTYPE;
8941: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8942: ncovv++; /* Varying variables without age */
8943: TvarV[ncovv]=Tvar[k];
8944: TvarVind[ncovv]=k;
8945: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8946: Fixed[k]= 1;
8947: Dummy[k]= 1;
8948: modell[k].maintype= VTYPE;
8949: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8950: ncovv++; /* Varying variables without age */
8951: TvarV[ncovv]=Tvar[k];
8952: TvarVind[ncovv]=k;
8953: }
1.227 brouard 8954: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8955: if(Tvard[k1][2] <=ncovcol){
8956: Fixed[k]= 0; /* or 2 ?*/
8957: Dummy[k]= 1;
8958: modell[k].maintype= FTYPE;
8959: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8960: ncovf++; /* Fixed variables without age */
8961: TvarF[ncovf]=Tvar[k];
8962: TvarFind[ncovf]=k;
8963: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8964: Fixed[k]= 1;
8965: Dummy[k]= 1;
8966: modell[k].maintype= VTYPE;
8967: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8968: ncovv++; /* Varying variables without age */
8969: TvarV[ncovv]=Tvar[k];
8970: TvarVind[ncovv]=k;
8971: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8972: Fixed[k]= 1;
8973: Dummy[k]= 1;
8974: modell[k].maintype= VTYPE;
8975: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8976: ncovv++; /* Varying variables without age */
8977: TvarV[ncovv]=Tvar[k];
8978: TvarVind[ncovv]=k;
8979: ncovv++; /* Varying variables without age */
8980: TvarV[ncovv]=Tvar[k];
8981: TvarVind[ncovv]=k;
8982: }
1.227 brouard 8983: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8984: if(Tvard[k1][2] <=ncovcol){
8985: Fixed[k]= 1;
8986: Dummy[k]= 1;
8987: modell[k].maintype= VTYPE;
8988: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8989: ncovv++; /* Varying variables without age */
8990: TvarV[ncovv]=Tvar[k];
8991: TvarVind[ncovv]=k;
8992: }else if(Tvard[k1][2] <=ncovcol+nqv){
8993: Fixed[k]= 1;
8994: Dummy[k]= 1;
8995: modell[k].maintype= VTYPE;
8996: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8997: ncovv++; /* Varying variables without age */
8998: TvarV[ncovv]=Tvar[k];
8999: TvarVind[ncovv]=k;
9000: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9001: Fixed[k]= 1;
9002: Dummy[k]= 0;
9003: modell[k].maintype= VTYPE;
9004: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9005: ncovv++; /* Varying variables without age */
9006: TvarV[ncovv]=Tvar[k];
9007: TvarVind[ncovv]=k;
9008: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9009: Fixed[k]= 1;
9010: Dummy[k]= 1;
9011: modell[k].maintype= VTYPE;
9012: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9013: ncovv++; /* Varying variables without age */
9014: TvarV[ncovv]=Tvar[k];
9015: TvarVind[ncovv]=k;
9016: }
1.227 brouard 9017: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9018: if(Tvard[k1][2] <=ncovcol){
9019: Fixed[k]= 1;
9020: Dummy[k]= 1;
9021: modell[k].maintype= VTYPE;
9022: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9023: ncovv++; /* Varying variables without age */
9024: TvarV[ncovv]=Tvar[k];
9025: TvarVind[ncovv]=k;
9026: }else if(Tvard[k1][2] <=ncovcol+nqv){
9027: Fixed[k]= 1;
9028: Dummy[k]= 1;
9029: modell[k].maintype= VTYPE;
9030: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9031: ncovv++; /* Varying variables without age */
9032: TvarV[ncovv]=Tvar[k];
9033: TvarVind[ncovv]=k;
9034: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9035: Fixed[k]= 1;
9036: Dummy[k]= 1;
9037: modell[k].maintype= VTYPE;
9038: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9039: ncovv++; /* Varying variables without age */
9040: TvarV[ncovv]=Tvar[k];
9041: TvarVind[ncovv]=k;
9042: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9043: Fixed[k]= 1;
9044: Dummy[k]= 1;
9045: modell[k].maintype= VTYPE;
9046: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9047: ncovv++; /* Varying variables without age */
9048: TvarV[ncovv]=Tvar[k];
9049: TvarVind[ncovv]=k;
9050: }
1.227 brouard 9051: }else{
1.240 brouard 9052: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9053: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9054: } /*end k1*/
1.225 brouard 9055: }else{
1.226 brouard 9056: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9057: 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 9058: }
1.227 brouard 9059: 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 9060: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9061: 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]);
9062: }
9063: /* Searching for doublons in the model */
9064: for(k1=1; k1<= cptcovt;k1++){
9065: for(k2=1; k2 <k1;k2++){
9066: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9067: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9068: if(Tvar[k1]==Tvar[k2]){
9069: 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]]);
9070: 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);
9071: return(1);
9072: }
9073: }else if (Typevar[k1] ==2){
9074: k3=Tposprod[k1];
9075: k4=Tposprod[k2];
9076: 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])) ){
9077: 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]]);
9078: 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);
9079: return(1);
9080: }
9081: }
1.227 brouard 9082: }
9083: }
1.225 brouard 9084: }
9085: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9086: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9087: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9088: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9089: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9090: /*endread:*/
1.225 brouard 9091: printf("Exiting decodemodel: ");
9092: return (1);
1.136 brouard 9093: }
9094:
1.169 brouard 9095: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9096: {/* Check ages at death */
1.136 brouard 9097: int i, m;
1.218 brouard 9098: int firstone=0;
9099:
1.136 brouard 9100: for (i=1; i<=imx; i++) {
9101: for(m=2; (m<= maxwav); m++) {
9102: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9103: anint[m][i]=9999;
1.216 brouard 9104: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9105: s[m][i]=-1;
1.136 brouard 9106: }
9107: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.259 ! brouard 9108: *nbwarn = *nbwarn + 1;
1.218 brouard 9109: if(firstone == 0){
9110: firstone=1;
1.259 ! brouard 9111: 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 believe in a death.\nOther similar cases in log file\n", *nbwarn,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 9112: }
1.259 ! brouard 9113: 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 believe in a death.\nOther similar cases in log file\n", *nbwarn,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
! 9114: /* s[m][i]=-1; */ /* Keeping the death status */
1.136 brouard 9115: }
9116: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9117: (*nberr)++;
1.259 ! brouard 9118: 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);
! 9119: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
! 9120: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9121: }
9122: }
9123: }
9124:
9125: for (i=1; i<=imx; i++) {
9126: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9127: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9128: 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 9129: if (s[m][i] >= nlstate+1) {
1.169 brouard 9130: if(agedc[i]>0){
9131: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9132: agev[m][i]=agedc[i];
1.214 brouard 9133: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9134: }else {
1.136 brouard 9135: if ((int)andc[i]!=9999){
9136: nbwarn++;
9137: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9138: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9139: agev[m][i]=-1;
9140: }
9141: }
1.169 brouard 9142: } /* agedc > 0 */
1.214 brouard 9143: } /* end if */
1.136 brouard 9144: else if(s[m][i] !=9){ /* Standard case, age in fractional
9145: years but with the precision of a month */
9146: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9147: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9148: agev[m][i]=1;
9149: else if(agev[m][i] < *agemin){
9150: *agemin=agev[m][i];
9151: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9152: }
9153: else if(agev[m][i] >*agemax){
9154: *agemax=agev[m][i];
1.156 brouard 9155: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9156: }
9157: /*agev[m][i]=anint[m][i]-annais[i];*/
9158: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9159: } /* en if 9*/
1.136 brouard 9160: else { /* =9 */
1.214 brouard 9161: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9162: agev[m][i]=1;
9163: s[m][i]=-1;
9164: }
9165: }
1.214 brouard 9166: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9167: agev[m][i]=1;
1.214 brouard 9168: else{
9169: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9170: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9171: agev[m][i]=0;
9172: }
9173: } /* End for lastpass */
9174: }
1.136 brouard 9175:
9176: for (i=1; i<=imx; i++) {
9177: for(m=firstpass; (m<=lastpass); m++){
9178: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9179: (*nberr)++;
1.136 brouard 9180: 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);
9181: 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);
9182: return 1;
9183: }
9184: }
9185: }
9186:
9187: /*for (i=1; i<=imx; i++){
9188: for (m=firstpass; (m<lastpass); m++){
9189: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9190: }
9191:
9192: }*/
9193:
9194:
1.139 brouard 9195: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9196: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9197:
9198: return (0);
1.164 brouard 9199: /* endread:*/
1.136 brouard 9200: printf("Exiting calandcheckages: ");
9201: return (1);
9202: }
9203:
1.172 brouard 9204: #if defined(_MSC_VER)
9205: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9206: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9207: //#include "stdafx.h"
9208: //#include <stdio.h>
9209: //#include <tchar.h>
9210: //#include <windows.h>
9211: //#include <iostream>
9212: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9213:
9214: LPFN_ISWOW64PROCESS fnIsWow64Process;
9215:
9216: BOOL IsWow64()
9217: {
9218: BOOL bIsWow64 = FALSE;
9219:
9220: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9221: // (HANDLE, PBOOL);
9222:
9223: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9224:
9225: HMODULE module = GetModuleHandle(_T("kernel32"));
9226: const char funcName[] = "IsWow64Process";
9227: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9228: GetProcAddress(module, funcName);
9229:
9230: if (NULL != fnIsWow64Process)
9231: {
9232: if (!fnIsWow64Process(GetCurrentProcess(),
9233: &bIsWow64))
9234: //throw std::exception("Unknown error");
9235: printf("Unknown error\n");
9236: }
9237: return bIsWow64 != FALSE;
9238: }
9239: #endif
1.177 brouard 9240:
1.191 brouard 9241: void syscompilerinfo(int logged)
1.167 brouard 9242: {
9243: /* #include "syscompilerinfo.h"*/
1.185 brouard 9244: /* command line Intel compiler 32bit windows, XP compatible:*/
9245: /* /GS /W3 /Gy
9246: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9247: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9248: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9249: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9250: */
9251: /* 64 bits */
1.185 brouard 9252: /*
9253: /GS /W3 /Gy
9254: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9255: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9256: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9257: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9258: /* Optimization are useless and O3 is slower than O2 */
9259: /*
9260: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9261: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9262: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9263: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9264: */
1.186 brouard 9265: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9266: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9267: /PDB:"visual studio
9268: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9269: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9270: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9271: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9272: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9273: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9274: uiAccess='false'"
9275: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9276: /NOLOGO /TLBID:1
9277: */
1.177 brouard 9278: #if defined __INTEL_COMPILER
1.178 brouard 9279: #if defined(__GNUC__)
9280: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9281: #endif
1.177 brouard 9282: #elif defined(__GNUC__)
1.179 brouard 9283: #ifndef __APPLE__
1.174 brouard 9284: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9285: #endif
1.177 brouard 9286: struct utsname sysInfo;
1.178 brouard 9287: int cross = CROSS;
9288: if (cross){
9289: printf("Cross-");
1.191 brouard 9290: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9291: }
1.174 brouard 9292: #endif
9293:
1.171 brouard 9294: #include <stdint.h>
1.178 brouard 9295:
1.191 brouard 9296: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9297: #if defined(__clang__)
1.191 brouard 9298: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9299: #endif
9300: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9301: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9302: #endif
9303: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9304: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9305: #endif
9306: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9307: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9308: #endif
9309: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9310: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9311: #endif
9312: #if defined(_MSC_VER)
1.191 brouard 9313: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9314: #endif
9315: #if defined(__PGI)
1.191 brouard 9316: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9317: #endif
9318: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9319: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9320: #endif
1.191 brouard 9321: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9322:
1.167 brouard 9323: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9324: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9325: // Windows (x64 and x86)
1.191 brouard 9326: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9327: #elif __unix__ // all unices, not all compilers
9328: // Unix
1.191 brouard 9329: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9330: #elif __linux__
9331: // linux
1.191 brouard 9332: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9333: #elif __APPLE__
1.174 brouard 9334: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9335: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9336: #endif
9337:
9338: /* __MINGW32__ */
9339: /* __CYGWIN__ */
9340: /* __MINGW64__ */
9341: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9342: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9343: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9344: /* _WIN64 // Defined for applications for Win64. */
9345: /* _M_X64 // Defined for compilations that target x64 processors. */
9346: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9347:
1.167 brouard 9348: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9349: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9350: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9351: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9352: #else
1.191 brouard 9353: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9354: #endif
9355:
1.169 brouard 9356: #if defined(__GNUC__)
9357: # if defined(__GNUC_PATCHLEVEL__)
9358: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9359: + __GNUC_MINOR__ * 100 \
9360: + __GNUC_PATCHLEVEL__)
9361: # else
9362: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9363: + __GNUC_MINOR__ * 100)
9364: # endif
1.174 brouard 9365: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9366: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9367:
9368: if (uname(&sysInfo) != -1) {
9369: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9370: 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 9371: }
9372: else
9373: perror("uname() error");
1.179 brouard 9374: //#ifndef __INTEL_COMPILER
9375: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9376: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9377: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9378: #endif
1.169 brouard 9379: #endif
1.172 brouard 9380:
9381: // void main()
9382: // {
1.169 brouard 9383: #if defined(_MSC_VER)
1.174 brouard 9384: if (IsWow64()){
1.191 brouard 9385: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9386: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9387: }
9388: else{
1.191 brouard 9389: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9390: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9391: }
1.172 brouard 9392: // printf("\nPress Enter to continue...");
9393: // getchar();
9394: // }
9395:
1.169 brouard 9396: #endif
9397:
1.167 brouard 9398:
1.219 brouard 9399: }
1.136 brouard 9400:
1.219 brouard 9401: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9402: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9403: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9404: /* double ftolpl = 1.e-10; */
1.180 brouard 9405: double age, agebase, agelim;
1.203 brouard 9406: double tot;
1.180 brouard 9407:
1.202 brouard 9408: strcpy(filerespl,"PL_");
9409: strcat(filerespl,fileresu);
9410: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9411: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9412: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9413: }
1.227 brouard 9414: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9415: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9416: pstamp(ficrespl);
1.203 brouard 9417: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9418: fprintf(ficrespl,"#Age ");
9419: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9420: fprintf(ficrespl,"\n");
1.180 brouard 9421:
1.219 brouard 9422: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9423:
1.219 brouard 9424: agebase=ageminpar;
9425: agelim=agemaxpar;
1.180 brouard 9426:
1.227 brouard 9427: /* i1=pow(2,ncoveff); */
1.234 brouard 9428: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9429: if (cptcovn < 1){i1=1;}
1.180 brouard 9430:
1.238 brouard 9431: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9432: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9433: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9434: continue;
1.235 brouard 9435:
1.238 brouard 9436: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9437: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9438: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9439: /* k=k+1; */
9440: /* to clean */
9441: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9442: fprintf(ficrespl,"#******");
9443: printf("#******");
9444: fprintf(ficlog,"#******");
9445: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9446: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9447: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9448: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9449: }
9450: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9451: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9452: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9453: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9454: }
9455: fprintf(ficrespl,"******\n");
9456: printf("******\n");
9457: fprintf(ficlog,"******\n");
9458: if(invalidvarcomb[k]){
9459: printf("\nCombination (%d) ignored because no case \n",k);
9460: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9461: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9462: continue;
9463: }
1.219 brouard 9464:
1.238 brouard 9465: fprintf(ficrespl,"#Age ");
9466: for(j=1;j<=cptcoveff;j++) {
9467: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9468: }
9469: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9470: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9471:
1.238 brouard 9472: for (age=agebase; age<=agelim; age++){
9473: /* for (age=agebase; age<=agebase; age++){ */
9474: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9475: fprintf(ficrespl,"%.0f ",age );
9476: for(j=1;j<=cptcoveff;j++)
9477: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9478: tot=0.;
9479: for(i=1; i<=nlstate;i++){
9480: tot += prlim[i][i];
9481: fprintf(ficrespl," %.5f", prlim[i][i]);
9482: }
9483: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9484: } /* Age */
9485: /* was end of cptcod */
9486: } /* cptcov */
9487: } /* nres */
1.219 brouard 9488: return 0;
1.180 brouard 9489: }
9490:
1.218 brouard 9491: 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){
9492: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9493:
9494: /* Computes the back prevalence limit for any combination of covariate values
9495: * at any age between ageminpar and agemaxpar
9496: */
1.235 brouard 9497: int i, j, k, i1, nres=0 ;
1.217 brouard 9498: /* double ftolpl = 1.e-10; */
9499: double age, agebase, agelim;
9500: double tot;
1.218 brouard 9501: /* double ***mobaverage; */
9502: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9503:
9504: strcpy(fileresplb,"PLB_");
9505: strcat(fileresplb,fileresu);
9506: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9507: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9508: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9509: }
9510: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9511: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9512: pstamp(ficresplb);
9513: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9514: fprintf(ficresplb,"#Age ");
9515: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9516: fprintf(ficresplb,"\n");
9517:
1.218 brouard 9518:
9519: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9520:
9521: agebase=ageminpar;
9522: agelim=agemaxpar;
9523:
9524:
1.227 brouard 9525: i1=pow(2,cptcoveff);
1.218 brouard 9526: if (cptcovn < 1){i1=1;}
1.227 brouard 9527:
1.238 brouard 9528: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9529: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9530: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9531: continue;
9532: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9533: fprintf(ficresplb,"#******");
9534: printf("#******");
9535: fprintf(ficlog,"#******");
9536: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9537: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9538: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9539: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9540: }
9541: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9542: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9543: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9544: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9545: }
9546: fprintf(ficresplb,"******\n");
9547: printf("******\n");
9548: fprintf(ficlog,"******\n");
9549: if(invalidvarcomb[k]){
9550: printf("\nCombination (%d) ignored because no cases \n",k);
9551: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9552: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9553: continue;
9554: }
1.218 brouard 9555:
1.238 brouard 9556: fprintf(ficresplb,"#Age ");
9557: for(j=1;j<=cptcoveff;j++) {
9558: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9559: }
9560: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9561: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9562:
9563:
1.238 brouard 9564: for (age=agebase; age<=agelim; age++){
9565: /* for (age=agebase; age<=agebase; age++){ */
9566: if(mobilavproj > 0){
9567: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9568: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9569: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9570: }else if (mobilavproj == 0){
9571: 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);
9572: 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);
9573: exit(1);
9574: }else{
9575: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9576: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9577: }
9578: fprintf(ficresplb,"%.0f ",age );
9579: for(j=1;j<=cptcoveff;j++)
9580: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9581: tot=0.;
9582: for(i=1; i<=nlstate;i++){
9583: tot += bprlim[i][i];
9584: fprintf(ficresplb," %.5f", bprlim[i][i]);
9585: }
9586: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9587: } /* Age */
9588: /* was end of cptcod */
1.255 brouard 9589: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9590: } /* end of any combination */
9591: } /* end of nres */
1.218 brouard 9592: /* hBijx(p, bage, fage); */
9593: /* fclose(ficrespijb); */
9594:
9595: return 0;
1.217 brouard 9596: }
1.218 brouard 9597:
1.180 brouard 9598: int hPijx(double *p, int bage, int fage){
9599: /*------------- h Pij x at various ages ------------*/
9600:
9601: int stepsize;
9602: int agelim;
9603: int hstepm;
9604: int nhstepm;
1.235 brouard 9605: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9606:
9607: double agedeb;
9608: double ***p3mat;
9609:
1.201 brouard 9610: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9611: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9612: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9613: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9614: }
9615: printf("Computing pij: result on file '%s' \n", filerespij);
9616: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9617:
9618: stepsize=(int) (stepm+YEARM-1)/YEARM;
9619: /*if (stepm<=24) stepsize=2;*/
9620:
9621: agelim=AGESUP;
9622: hstepm=stepsize*YEARM; /* Every year of age */
9623: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9624:
1.180 brouard 9625: /* hstepm=1; aff par mois*/
9626: pstamp(ficrespij);
9627: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9628: i1= pow(2,cptcoveff);
1.218 brouard 9629: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9630: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9631: /* k=k+1; */
1.235 brouard 9632: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9633: for(k=1; k<=i1;k++){
1.253 brouard 9634: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9635: continue;
1.183 brouard 9636: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9637: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9638: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9639: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9640: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9641: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9642: }
1.183 brouard 9643: fprintf(ficrespij,"******\n");
9644:
9645: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9646: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9647: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9648:
9649: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9650:
1.183 brouard 9651: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9652: oldm=oldms;savm=savms;
1.235 brouard 9653: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9654: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9655: for(i=1; i<=nlstate;i++)
9656: for(j=1; j<=nlstate+ndeath;j++)
9657: fprintf(ficrespij," %1d-%1d",i,j);
9658: fprintf(ficrespij,"\n");
9659: for (h=0; h<=nhstepm; h++){
9660: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9661: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9662: for(i=1; i<=nlstate;i++)
9663: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9664: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9665: fprintf(ficrespij,"\n");
9666: }
1.183 brouard 9667: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9668: fprintf(ficrespij,"\n");
9669: }
1.180 brouard 9670: /*}*/
9671: }
1.218 brouard 9672: return 0;
1.180 brouard 9673: }
1.218 brouard 9674:
9675: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9676: /*------------- h Bij x at various ages ------------*/
9677:
9678: int stepsize;
1.218 brouard 9679: /* int agelim; */
9680: int ageminl;
1.217 brouard 9681: int hstepm;
9682: int nhstepm;
1.238 brouard 9683: int h, i, i1, j, k, nres;
1.218 brouard 9684:
1.217 brouard 9685: double agedeb;
9686: double ***p3mat;
1.218 brouard 9687:
9688: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9689: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9690: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9691: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9692: }
9693: printf("Computing pij back: result on file '%s' \n", filerespijb);
9694: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9695:
9696: stepsize=(int) (stepm+YEARM-1)/YEARM;
9697: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9698:
1.218 brouard 9699: /* agelim=AGESUP; */
9700: ageminl=30;
9701: hstepm=stepsize*YEARM; /* Every year of age */
9702: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9703:
9704: /* hstepm=1; aff par mois*/
9705: pstamp(ficrespijb);
1.255 brouard 9706: 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 9707: i1= pow(2,cptcoveff);
1.218 brouard 9708: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9709: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9710: /* k=k+1; */
1.238 brouard 9711: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9712: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9713: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9714: continue;
9715: fprintf(ficrespijb,"\n#****** ");
9716: for(j=1;j<=cptcoveff;j++)
9717: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9718: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9719: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9720: }
9721: fprintf(ficrespijb,"******\n");
9722: if(invalidvarcomb[k]){
9723: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9724: continue;
9725: }
9726:
9727: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9728: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9729: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9730: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9731: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9732:
9733: /* nhstepm=nhstepm*YEARM; aff par mois*/
9734:
9735: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9736: /* oldm=oldms;savm=savms; */
9737: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9738: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9739: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9740: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9741: for(i=1; i<=nlstate;i++)
9742: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9743: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9744: fprintf(ficrespijb,"\n");
1.238 brouard 9745: for (h=0; h<=nhstepm; h++){
9746: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9747: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9748: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9749: for(i=1; i<=nlstate;i++)
9750: for(j=1; j<=nlstate+ndeath;j++)
9751: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9752: fprintf(ficrespijb,"\n");
9753: }
9754: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9755: fprintf(ficrespijb,"\n");
9756: } /* end age deb */
9757: } /* end combination */
9758: } /* end nres */
1.218 brouard 9759: return 0;
9760: } /* hBijx */
1.217 brouard 9761:
1.180 brouard 9762:
1.136 brouard 9763: /***********************************************/
9764: /**************** Main Program *****************/
9765: /***********************************************/
9766:
9767: int main(int argc, char *argv[])
9768: {
9769: #ifdef GSL
9770: const gsl_multimin_fminimizer_type *T;
9771: size_t iteri = 0, it;
9772: int rval = GSL_CONTINUE;
9773: int status = GSL_SUCCESS;
9774: double ssval;
9775: #endif
9776: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9777: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9778: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9779: int jj, ll, li, lj, lk;
1.136 brouard 9780: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9781: int num_filled;
1.136 brouard 9782: int itimes;
9783: int NDIM=2;
9784: int vpopbased=0;
1.235 brouard 9785: int nres=0;
1.258 brouard 9786: int endishere=0;
1.136 brouard 9787:
1.164 brouard 9788: char ca[32], cb[32];
1.136 brouard 9789: /* FILE *fichtm; *//* Html File */
9790: /* FILE *ficgp;*/ /*Gnuplot File */
9791: struct stat info;
1.191 brouard 9792: double agedeb=0.;
1.194 brouard 9793:
9794: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9795: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9796:
1.165 brouard 9797: double fret;
1.191 brouard 9798: double dum=0.; /* Dummy variable */
1.136 brouard 9799: double ***p3mat;
1.218 brouard 9800: /* double ***mobaverage; */
1.164 brouard 9801:
9802: char line[MAXLINE];
1.197 brouard 9803: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9804:
1.234 brouard 9805: char modeltemp[MAXLINE];
1.230 brouard 9806: char resultline[MAXLINE];
9807:
1.136 brouard 9808: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9809: char *tok, *val; /* pathtot */
1.136 brouard 9810: int firstobs=1, lastobs=10;
1.195 brouard 9811: int c, h , cpt, c2;
1.191 brouard 9812: int jl=0;
9813: int i1, j1, jk, stepsize=0;
1.194 brouard 9814: int count=0;
9815:
1.164 brouard 9816: int *tab;
1.136 brouard 9817: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9818: int backcast=0;
1.136 brouard 9819: int mobilav=0,popforecast=0;
1.191 brouard 9820: int hstepm=0, nhstepm=0;
1.136 brouard 9821: int agemortsup;
9822: float sumlpop=0.;
9823: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9824: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9825:
1.191 brouard 9826: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9827: double ftolpl=FTOL;
9828: double **prlim;
1.217 brouard 9829: double **bprlim;
1.136 brouard 9830: double ***param; /* Matrix of parameters */
1.251 brouard 9831: double ***paramstart; /* Matrix of starting parameter values */
9832: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9833: double **matcov; /* Matrix of covariance */
1.203 brouard 9834: double **hess; /* Hessian matrix */
1.136 brouard 9835: double ***delti3; /* Scale */
9836: double *delti; /* Scale */
9837: double ***eij, ***vareij;
9838: double **varpl; /* Variances of prevalence limits by age */
9839: double *epj, vepp;
1.164 brouard 9840:
1.136 brouard 9841: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9842: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9843:
1.136 brouard 9844: double **ximort;
1.145 brouard 9845: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9846: int *dcwave;
9847:
1.164 brouard 9848: char z[1]="c";
1.136 brouard 9849:
9850: /*char *strt;*/
9851: char strtend[80];
1.126 brouard 9852:
1.164 brouard 9853:
1.126 brouard 9854: /* setlocale (LC_ALL, ""); */
9855: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9856: /* textdomain (PACKAGE); */
9857: /* setlocale (LC_CTYPE, ""); */
9858: /* setlocale (LC_MESSAGES, ""); */
9859:
9860: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9861: rstart_time = time(NULL);
9862: /* (void) gettimeofday(&start_time,&tzp);*/
9863: start_time = *localtime(&rstart_time);
1.126 brouard 9864: curr_time=start_time;
1.157 brouard 9865: /*tml = *localtime(&start_time.tm_sec);*/
9866: /* strcpy(strstart,asctime(&tml)); */
9867: strcpy(strstart,asctime(&start_time));
1.126 brouard 9868:
9869: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9870: /* tp.tm_sec = tp.tm_sec +86400; */
9871: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9872: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9873: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9874: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9875: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9876: /* strt=asctime(&tmg); */
9877: /* printf("Time(after) =%s",strstart); */
9878: /* (void) time (&time_value);
9879: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9880: * tm = *localtime(&time_value);
9881: * strstart=asctime(&tm);
9882: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9883: */
9884:
9885: nberr=0; /* Number of errors and warnings */
9886: nbwarn=0;
1.184 brouard 9887: #ifdef WIN32
9888: _getcwd(pathcd, size);
9889: #else
1.126 brouard 9890: getcwd(pathcd, size);
1.184 brouard 9891: #endif
1.191 brouard 9892: syscompilerinfo(0);
1.196 brouard 9893: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9894: if(argc <=1){
9895: printf("\nEnter the parameter file name: ");
1.205 brouard 9896: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9897: printf("ERROR Empty parameter file name\n");
9898: goto end;
9899: }
1.126 brouard 9900: i=strlen(pathr);
9901: if(pathr[i-1]=='\n')
9902: pathr[i-1]='\0';
1.156 brouard 9903: i=strlen(pathr);
1.205 brouard 9904: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9905: pathr[i-1]='\0';
1.205 brouard 9906: }
9907: i=strlen(pathr);
9908: if( i==0 ){
9909: printf("ERROR Empty parameter file name\n");
9910: goto end;
9911: }
9912: for (tok = pathr; tok != NULL; ){
1.126 brouard 9913: printf("Pathr |%s|\n",pathr);
9914: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9915: printf("val= |%s| pathr=%s\n",val,pathr);
9916: strcpy (pathtot, val);
9917: if(pathr[0] == '\0') break; /* Dirty */
9918: }
9919: }
9920: else{
9921: strcpy(pathtot,argv[1]);
9922: }
9923: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9924: /*cygwin_split_path(pathtot,path,optionfile);
9925: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9926: /* cutv(path,optionfile,pathtot,'\\');*/
9927:
9928: /* Split argv[0], imach program to get pathimach */
9929: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9930: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9931: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9932: /* strcpy(pathimach,argv[0]); */
9933: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9934: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9935: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9936: #ifdef WIN32
9937: _chdir(path); /* Can be a relative path */
9938: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9939: #else
1.126 brouard 9940: chdir(path); /* Can be a relative path */
1.184 brouard 9941: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9942: #endif
9943: printf("Current directory %s!\n",pathcd);
1.126 brouard 9944: strcpy(command,"mkdir ");
9945: strcat(command,optionfilefiname);
9946: if((outcmd=system(command)) != 0){
1.169 brouard 9947: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9948: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9949: /* fclose(ficlog); */
9950: /* exit(1); */
9951: }
9952: /* if((imk=mkdir(optionfilefiname))<0){ */
9953: /* perror("mkdir"); */
9954: /* } */
9955:
9956: /*-------- arguments in the command line --------*/
9957:
1.186 brouard 9958: /* Main Log file */
1.126 brouard 9959: strcat(filelog, optionfilefiname);
9960: strcat(filelog,".log"); /* */
9961: if((ficlog=fopen(filelog,"w"))==NULL) {
9962: printf("Problem with logfile %s\n",filelog);
9963: goto end;
9964: }
9965: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9966: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9967: fprintf(ficlog,"\nEnter the parameter file name: \n");
9968: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9969: path=%s \n\
9970: optionfile=%s\n\
9971: optionfilext=%s\n\
1.156 brouard 9972: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9973:
1.197 brouard 9974: syscompilerinfo(1);
1.167 brouard 9975:
1.126 brouard 9976: printf("Local time (at start):%s",strstart);
9977: fprintf(ficlog,"Local time (at start): %s",strstart);
9978: fflush(ficlog);
9979: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9980: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9981:
9982: /* */
9983: strcpy(fileres,"r");
9984: strcat(fileres, optionfilefiname);
1.201 brouard 9985: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9986: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9987: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9988:
1.186 brouard 9989: /* Main ---------arguments file --------*/
1.126 brouard 9990:
9991: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9992: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9993: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9994: fflush(ficlog);
1.149 brouard 9995: /* goto end; */
9996: exit(70);
1.126 brouard 9997: }
9998:
9999:
10000:
10001: strcpy(filereso,"o");
1.201 brouard 10002: strcat(filereso,fileresu);
1.126 brouard 10003: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10004: printf("Problem with Output resultfile: %s\n", filereso);
10005: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10006: fflush(ficlog);
10007: goto end;
10008: }
10009:
10010: /* Reads comments: lines beginning with '#' */
10011: numlinepar=0;
1.197 brouard 10012:
10013: /* First parameter line */
10014: while(fgets(line, MAXLINE, ficpar)) {
10015: /* If line starts with a # it is a comment */
10016: if (line[0] == '#') {
10017: numlinepar++;
10018: fputs(line,stdout);
10019: fputs(line,ficparo);
10020: fputs(line,ficlog);
10021: continue;
10022: }else
10023: break;
10024: }
10025: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10026: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10027: if (num_filled != 5) {
10028: printf("Should be 5 parameters\n");
10029: }
1.126 brouard 10030: numlinepar++;
1.197 brouard 10031: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10032: }
10033: /* Second parameter line */
10034: while(fgets(line, MAXLINE, ficpar)) {
10035: /* If line starts with a # it is a comment */
10036: if (line[0] == '#') {
10037: numlinepar++;
10038: fputs(line,stdout);
10039: fputs(line,ficparo);
10040: fputs(line,ficlog);
10041: continue;
10042: }else
10043: break;
10044: }
1.223 brouard 10045: 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", \
10046: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10047: if (num_filled != 11) {
10048: 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 10049: printf("but line=%s\n",line);
1.197 brouard 10050: }
1.223 brouard 10051: 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 10052: }
1.203 brouard 10053: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10054: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10055: /* Third parameter line */
10056: while(fgets(line, MAXLINE, ficpar)) {
10057: /* If line starts with a # it is a comment */
10058: if (line[0] == '#') {
10059: numlinepar++;
10060: fputs(line,stdout);
10061: fputs(line,ficparo);
10062: fputs(line,ficlog);
10063: continue;
10064: }else
10065: break;
10066: }
1.201 brouard 10067: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10068: if (num_filled == 0)
10069: model[0]='\0';
10070: else if (num_filled != 1){
1.197 brouard 10071: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10072: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10073: model[0]='\0';
10074: goto end;
10075: }
10076: else{
10077: if (model[0]=='+'){
10078: for(i=1; i<=strlen(model);i++)
10079: modeltemp[i-1]=model[i];
1.201 brouard 10080: strcpy(model,modeltemp);
1.197 brouard 10081: }
10082: }
1.199 brouard 10083: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10084: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10085: }
10086: /* 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); */
10087: /* numlinepar=numlinepar+3; /\* In general *\/ */
10088: /* 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 10089: 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);
10090: 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 10091: fflush(ficlog);
1.190 brouard 10092: /* if(model[0]=='#'|| model[0]== '\0'){ */
10093: if(model[0]=='#'){
1.187 brouard 10094: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10095: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10096: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10097: if(mle != -1){
10098: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10099: exit(1);
10100: }
10101: }
1.126 brouard 10102: while((c=getc(ficpar))=='#' && c!= EOF){
10103: ungetc(c,ficpar);
10104: fgets(line, MAXLINE, ficpar);
10105: numlinepar++;
1.195 brouard 10106: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10107: z[0]=line[1];
10108: }
10109: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10110: fputs(line, stdout);
10111: //puts(line);
1.126 brouard 10112: fputs(line,ficparo);
10113: fputs(line,ficlog);
10114: }
10115: ungetc(c,ficpar);
10116:
10117:
1.145 brouard 10118: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10119: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10120: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10121: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10122: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10123: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10124: v1+v2*age+v2*v3 makes cptcovn = 3
10125: */
10126: if (strlen(model)>1)
1.187 brouard 10127: 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 10128: else
1.187 brouard 10129: ncovmodel=2; /* Constant and age */
1.133 brouard 10130: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10131: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10132: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10133: 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);
10134: 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);
10135: fflush(stdout);
10136: fclose (ficlog);
10137: goto end;
10138: }
1.126 brouard 10139: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10140: delti=delti3[1][1];
10141: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10142: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10143: /* We could also provide initial parameters values giving by simple logistic regression
10144: * only one way, that is without matrix product. We will have nlstate maximizations */
10145: /* for(i=1;i<nlstate;i++){ */
10146: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10147: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10148: /* } */
1.126 brouard 10149: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10150: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10151: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10152: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10153: fclose (ficparo);
10154: fclose (ficlog);
10155: goto end;
10156: exit(0);
1.220 brouard 10157: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10158: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10159: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10160: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10161: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10162: matcov=matrix(1,npar,1,npar);
1.203 brouard 10163: hess=matrix(1,npar,1,npar);
1.220 brouard 10164: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10165: /* Read guessed parameters */
1.126 brouard 10166: /* Reads comments: lines beginning with '#' */
10167: while((c=getc(ficpar))=='#' && c!= EOF){
10168: ungetc(c,ficpar);
10169: fgets(line, MAXLINE, ficpar);
10170: numlinepar++;
1.141 brouard 10171: fputs(line,stdout);
1.126 brouard 10172: fputs(line,ficparo);
10173: fputs(line,ficlog);
10174: }
10175: ungetc(c,ficpar);
10176:
10177: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10178: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10179: for(i=1; i <=nlstate; i++){
1.234 brouard 10180: j=0;
1.126 brouard 10181: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10182: if(jj==i) continue;
10183: j++;
10184: fscanf(ficpar,"%1d%1d",&i1,&j1);
10185: if ((i1 != i) || (j1 != jj)){
10186: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10187: It might be a problem of design; if ncovcol and the model are correct\n \
10188: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10189: exit(1);
10190: }
10191: fprintf(ficparo,"%1d%1d",i1,j1);
10192: if(mle==1)
10193: printf("%1d%1d",i,jj);
10194: fprintf(ficlog,"%1d%1d",i,jj);
10195: for(k=1; k<=ncovmodel;k++){
10196: fscanf(ficpar," %lf",¶m[i][j][k]);
10197: if(mle==1){
10198: printf(" %lf",param[i][j][k]);
10199: fprintf(ficlog," %lf",param[i][j][k]);
10200: }
10201: else
10202: fprintf(ficlog," %lf",param[i][j][k]);
10203: fprintf(ficparo," %lf",param[i][j][k]);
10204: }
10205: fscanf(ficpar,"\n");
10206: numlinepar++;
10207: if(mle==1)
10208: printf("\n");
10209: fprintf(ficlog,"\n");
10210: fprintf(ficparo,"\n");
1.126 brouard 10211: }
10212: }
10213: fflush(ficlog);
1.234 brouard 10214:
1.251 brouard 10215: /* Reads parameters values */
1.126 brouard 10216: p=param[1][1];
1.251 brouard 10217: pstart=paramstart[1][1];
1.126 brouard 10218:
10219: /* Reads comments: lines beginning with '#' */
10220: while((c=getc(ficpar))=='#' && c!= EOF){
10221: ungetc(c,ficpar);
10222: fgets(line, MAXLINE, ficpar);
10223: numlinepar++;
1.141 brouard 10224: fputs(line,stdout);
1.126 brouard 10225: fputs(line,ficparo);
10226: fputs(line,ficlog);
10227: }
10228: ungetc(c,ficpar);
10229:
10230: for(i=1; i <=nlstate; i++){
10231: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10232: fscanf(ficpar,"%1d%1d",&i1,&j1);
10233: if ( (i1-i) * (j1-j) != 0){
10234: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10235: exit(1);
10236: }
10237: printf("%1d%1d",i,j);
10238: fprintf(ficparo,"%1d%1d",i1,j1);
10239: fprintf(ficlog,"%1d%1d",i1,j1);
10240: for(k=1; k<=ncovmodel;k++){
10241: fscanf(ficpar,"%le",&delti3[i][j][k]);
10242: printf(" %le",delti3[i][j][k]);
10243: fprintf(ficparo," %le",delti3[i][j][k]);
10244: fprintf(ficlog," %le",delti3[i][j][k]);
10245: }
10246: fscanf(ficpar,"\n");
10247: numlinepar++;
10248: printf("\n");
10249: fprintf(ficparo,"\n");
10250: fprintf(ficlog,"\n");
1.126 brouard 10251: }
10252: }
10253: fflush(ficlog);
1.234 brouard 10254:
1.145 brouard 10255: /* Reads covariance matrix */
1.126 brouard 10256: delti=delti3[1][1];
1.220 brouard 10257:
10258:
1.126 brouard 10259: /* 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 10260:
1.126 brouard 10261: /* Reads comments: lines beginning with '#' */
10262: while((c=getc(ficpar))=='#' && c!= EOF){
10263: ungetc(c,ficpar);
10264: fgets(line, MAXLINE, ficpar);
10265: numlinepar++;
1.141 brouard 10266: fputs(line,stdout);
1.126 brouard 10267: fputs(line,ficparo);
10268: fputs(line,ficlog);
10269: }
10270: ungetc(c,ficpar);
1.220 brouard 10271:
1.126 brouard 10272: matcov=matrix(1,npar,1,npar);
1.203 brouard 10273: hess=matrix(1,npar,1,npar);
1.131 brouard 10274: for(i=1; i <=npar; i++)
10275: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10276:
1.194 brouard 10277: /* Scans npar lines */
1.126 brouard 10278: for(i=1; i <=npar; i++){
1.226 brouard 10279: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10280: if(count != 3){
1.226 brouard 10281: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10282: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10283: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10284: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10285: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10286: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10287: exit(1);
1.220 brouard 10288: }else{
1.226 brouard 10289: if(mle==1)
10290: printf("%1d%1d%d",i1,j1,jk);
10291: }
10292: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10293: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10294: for(j=1; j <=i; j++){
1.226 brouard 10295: fscanf(ficpar," %le",&matcov[i][j]);
10296: if(mle==1){
10297: printf(" %.5le",matcov[i][j]);
10298: }
10299: fprintf(ficlog," %.5le",matcov[i][j]);
10300: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10301: }
10302: fscanf(ficpar,"\n");
10303: numlinepar++;
10304: if(mle==1)
1.220 brouard 10305: printf("\n");
1.126 brouard 10306: fprintf(ficlog,"\n");
10307: fprintf(ficparo,"\n");
10308: }
1.194 brouard 10309: /* End of read covariance matrix npar lines */
1.126 brouard 10310: for(i=1; i <=npar; i++)
10311: for(j=i+1;j<=npar;j++)
1.226 brouard 10312: matcov[i][j]=matcov[j][i];
1.126 brouard 10313:
10314: if(mle==1)
10315: printf("\n");
10316: fprintf(ficlog,"\n");
10317:
10318: fflush(ficlog);
10319:
10320: /*-------- Rewriting parameter file ----------*/
10321: strcpy(rfileres,"r"); /* "Rparameterfile */
10322: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10323: strcat(rfileres,"."); /* */
10324: strcat(rfileres,optionfilext); /* Other files have txt extension */
10325: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10326: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10327: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10328: }
10329: fprintf(ficres,"#%s\n",version);
10330: } /* End of mle != -3 */
1.218 brouard 10331:
1.186 brouard 10332: /* Main data
10333: */
1.126 brouard 10334: n= lastobs;
10335: num=lvector(1,n);
10336: moisnais=vector(1,n);
10337: annais=vector(1,n);
10338: moisdc=vector(1,n);
10339: andc=vector(1,n);
1.220 brouard 10340: weight=vector(1,n);
1.126 brouard 10341: agedc=vector(1,n);
10342: cod=ivector(1,n);
1.220 brouard 10343: for(i=1;i<=n;i++){
1.234 brouard 10344: num[i]=0;
10345: moisnais[i]=0;
10346: annais[i]=0;
10347: moisdc[i]=0;
10348: andc[i]=0;
10349: agedc[i]=0;
10350: cod[i]=0;
10351: weight[i]=1.0; /* Equal weights, 1 by default */
10352: }
1.126 brouard 10353: mint=matrix(1,maxwav,1,n);
10354: anint=matrix(1,maxwav,1,n);
1.131 brouard 10355: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10356: tab=ivector(1,NCOVMAX);
1.144 brouard 10357: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10358: 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 10359:
1.136 brouard 10360: /* Reads data from file datafile */
10361: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10362: goto end;
10363:
10364: /* Calculation of the number of parameters from char model */
1.234 brouard 10365: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10366: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10367: k=3 V4 Tvar[k=3]= 4 (from V4)
10368: k=2 V1 Tvar[k=2]= 1 (from V1)
10369: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10370: */
10371:
10372: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10373: TvarsDind=ivector(1,NCOVMAX); /* */
10374: TvarsD=ivector(1,NCOVMAX); /* */
10375: TvarsQind=ivector(1,NCOVMAX); /* */
10376: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10377: TvarF=ivector(1,NCOVMAX); /* */
10378: TvarFind=ivector(1,NCOVMAX); /* */
10379: TvarV=ivector(1,NCOVMAX); /* */
10380: TvarVind=ivector(1,NCOVMAX); /* */
10381: TvarA=ivector(1,NCOVMAX); /* */
10382: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10383: TvarFD=ivector(1,NCOVMAX); /* */
10384: TvarFDind=ivector(1,NCOVMAX); /* */
10385: TvarFQ=ivector(1,NCOVMAX); /* */
10386: TvarFQind=ivector(1,NCOVMAX); /* */
10387: TvarVD=ivector(1,NCOVMAX); /* */
10388: TvarVDind=ivector(1,NCOVMAX); /* */
10389: TvarVQ=ivector(1,NCOVMAX); /* */
10390: TvarVQind=ivector(1,NCOVMAX); /* */
10391:
1.230 brouard 10392: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10393: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10394: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10395: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10396: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10397: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10398: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10399: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10400: */
10401: /* For model-covariate k tells which data-covariate to use but
10402: because this model-covariate is a construction we invent a new column
10403: ncovcol + k1
10404: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10405: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10406: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10407: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10408: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10409: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10410: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10411: */
1.145 brouard 10412: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10413: 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 10414: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10415: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10416: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10417: 4 covariates (3 plus signs)
10418: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10419: */
1.230 brouard 10420: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10421: * individual dummy, fixed or varying:
10422: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10423: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10424: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10425: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10426: * Tmodelind[1]@9={9,0,3,2,}*/
10427: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10428: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10429: * individual quantitative, fixed or varying:
10430: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10431: * 3, 1, 0, 0, 0, 0, 0, 0},
10432: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10433: /* Main decodemodel */
10434:
1.187 brouard 10435:
1.223 brouard 10436: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10437: goto end;
10438:
1.137 brouard 10439: if((double)(lastobs-imx)/(double)imx > 1.10){
10440: nbwarn++;
10441: 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);
10442: 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);
10443: }
1.136 brouard 10444: /* if(mle==1){*/
1.137 brouard 10445: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10446: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10447: }
10448:
10449: /*-calculation of age at interview from date of interview and age at death -*/
10450: agev=matrix(1,maxwav,1,imx);
10451:
10452: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10453: goto end;
10454:
1.126 brouard 10455:
1.136 brouard 10456: agegomp=(int)agemin;
10457: free_vector(moisnais,1,n);
10458: free_vector(annais,1,n);
1.126 brouard 10459: /* free_matrix(mint,1,maxwav,1,n);
10460: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10461: /* free_vector(moisdc,1,n); */
10462: /* free_vector(andc,1,n); */
1.145 brouard 10463: /* */
10464:
1.126 brouard 10465: wav=ivector(1,imx);
1.214 brouard 10466: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10467: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10468: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10469: 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.*/
10470: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10471: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10472:
10473: /* Concatenates waves */
1.214 brouard 10474: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10475: Death is a valid wave (if date is known).
10476: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10477: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10478: and mw[mi+1][i]. dh depends on stepm.
10479: */
10480:
1.126 brouard 10481: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10482: /* Concatenates waves */
1.145 brouard 10483:
1.215 brouard 10484: free_vector(moisdc,1,n);
10485: free_vector(andc,1,n);
10486:
1.126 brouard 10487: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10488: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10489: ncodemax[1]=1;
1.145 brouard 10490: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10491: cptcoveff=0;
1.220 brouard 10492: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10493: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10494: }
10495:
10496: ncovcombmax=pow(2,cptcoveff);
10497: invalidvarcomb=ivector(1, ncovcombmax);
10498: for(i=1;i<ncovcombmax;i++)
10499: invalidvarcomb[i]=0;
10500:
1.211 brouard 10501: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10502: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10503: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10504:
1.200 brouard 10505: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10506: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10507: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10508: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10509: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10510: * (currently 0 or 1) in the data.
10511: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10512: * corresponding modality (h,j).
10513: */
10514:
1.145 brouard 10515: h=0;
10516: /*if (cptcovn > 0) */
1.126 brouard 10517: m=pow(2,cptcoveff);
10518:
1.144 brouard 10519: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10520: * For k=4 covariates, h goes from 1 to m=2**k
10521: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10522: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10523: * h\k 1 2 3 4
1.143 brouard 10524: *______________________________
10525: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10526: * 2 2 1 1 1
10527: * 3 i=2 1 2 1 1
10528: * 4 2 2 1 1
10529: * 5 i=3 1 i=2 1 2 1
10530: * 6 2 1 2 1
10531: * 7 i=4 1 2 2 1
10532: * 8 2 2 2 1
1.197 brouard 10533: * 9 i=5 1 i=3 1 i=2 1 2
10534: * 10 2 1 1 2
10535: * 11 i=6 1 2 1 2
10536: * 12 2 2 1 2
10537: * 13 i=7 1 i=4 1 2 2
10538: * 14 2 1 2 2
10539: * 15 i=8 1 2 2 2
10540: * 16 2 2 2 2
1.143 brouard 10541: */
1.212 brouard 10542: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10543: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10544: * and the value of each covariate?
10545: * V1=1, V2=1, V3=2, V4=1 ?
10546: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10547: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10548: * In order to get the real value in the data, we use nbcode
10549: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10550: * We are keeping this crazy system in order to be able (in the future?)
10551: * to have more than 2 values (0 or 1) for a covariate.
10552: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10553: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10554: * bbbbbbbb
10555: * 76543210
10556: * h-1 00000101 (6-1=5)
1.219 brouard 10557: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10558: * &
10559: * 1 00000001 (1)
1.219 brouard 10560: * 00000000 = 1 & ((h-1) >> (k-1))
10561: * +1= 00000001 =1
1.211 brouard 10562: *
10563: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10564: * h' 1101 =2^3+2^2+0x2^1+2^0
10565: * >>k' 11
10566: * & 00000001
10567: * = 00000001
10568: * +1 = 00000010=2 = codtabm(14,3)
10569: * Reverse h=6 and m=16?
10570: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10571: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10572: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10573: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10574: * V3=decodtabm(14,3,2**4)=2
10575: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10576: *(h-1) >> (j-1) 0011 =13 >> 2
10577: * &1 000000001
10578: * = 000000001
10579: * +1= 000000010 =2
10580: * 2211
10581: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10582: * V3=2
1.220 brouard 10583: * codtabm and decodtabm are identical
1.211 brouard 10584: */
10585:
1.145 brouard 10586:
10587: free_ivector(Ndum,-1,NCOVMAX);
10588:
10589:
1.126 brouard 10590:
1.186 brouard 10591: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10592: strcpy(optionfilegnuplot,optionfilefiname);
10593: if(mle==-3)
1.201 brouard 10594: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10595: strcat(optionfilegnuplot,".gp");
10596:
10597: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10598: printf("Problem with file %s",optionfilegnuplot);
10599: }
10600: else{
1.204 brouard 10601: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10602: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10603: //fprintf(ficgp,"set missing 'NaNq'\n");
10604: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10605: }
10606: /* fclose(ficgp);*/
1.186 brouard 10607:
10608:
10609: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10610:
10611: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10612: if(mle==-3)
1.201 brouard 10613: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10614: strcat(optionfilehtm,".htm");
10615: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10616: printf("Problem with %s \n",optionfilehtm);
10617: exit(0);
1.126 brouard 10618: }
10619:
10620: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10621: strcat(optionfilehtmcov,"-cov.htm");
10622: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10623: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10624: }
10625: else{
10626: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10627: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10628: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10629: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10630: }
10631:
1.213 brouard 10632: 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 10633: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10634: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10635: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10636: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10637: \n\
10638: <hr size=\"2\" color=\"#EC5E5E\">\
10639: <ul><li><h4>Parameter files</h4>\n\
10640: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10641: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10642: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10643: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10644: - Date and time at start: %s</ul>\n",\
10645: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10646: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10647: fileres,fileres,\
10648: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10649: fflush(fichtm);
10650:
10651: strcpy(pathr,path);
10652: strcat(pathr,optionfilefiname);
1.184 brouard 10653: #ifdef WIN32
10654: _chdir(optionfilefiname); /* Move to directory named optionfile */
10655: #else
1.126 brouard 10656: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10657: #endif
10658:
1.126 brouard 10659:
1.220 brouard 10660: /* Calculates basic frequencies. Computes observed prevalence at single age
10661: and for any valid combination of covariates
1.126 brouard 10662: and prints on file fileres'p'. */
1.251 brouard 10663: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10664: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10665:
10666: fprintf(fichtm,"\n");
10667: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10668: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10669: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10670: imx,agemin,agemax,jmin,jmax,jmean);
10671: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10672: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10673: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10674: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10675: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10676:
1.126 brouard 10677: /* For Powell, parameters are in a vector p[] starting at p[1]
10678: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10679: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10680:
10681: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10682: /* For mortality only */
1.126 brouard 10683: if (mle==-3){
1.136 brouard 10684: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10685: for(i=1;i<=NDIM;i++)
10686: for(j=1;j<=NDIM;j++)
10687: ximort[i][j]=0.;
1.186 brouard 10688: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10689: cens=ivector(1,n);
10690: ageexmed=vector(1,n);
10691: agecens=vector(1,n);
10692: dcwave=ivector(1,n);
1.223 brouard 10693:
1.126 brouard 10694: for (i=1; i<=imx; i++){
10695: dcwave[i]=-1;
10696: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10697: if (s[m][i]>nlstate) {
10698: dcwave[i]=m;
10699: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10700: break;
10701: }
1.126 brouard 10702: }
1.226 brouard 10703:
1.126 brouard 10704: for (i=1; i<=imx; i++) {
10705: if (wav[i]>0){
1.226 brouard 10706: ageexmed[i]=agev[mw[1][i]][i];
10707: j=wav[i];
10708: agecens[i]=1.;
10709:
10710: if (ageexmed[i]> 1 && wav[i] > 0){
10711: agecens[i]=agev[mw[j][i]][i];
10712: cens[i]= 1;
10713: }else if (ageexmed[i]< 1)
10714: cens[i]= -1;
10715: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10716: cens[i]=0 ;
1.126 brouard 10717: }
10718: else cens[i]=-1;
10719: }
10720:
10721: for (i=1;i<=NDIM;i++) {
10722: for (j=1;j<=NDIM;j++)
1.226 brouard 10723: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10724: }
10725:
1.145 brouard 10726: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10727: /*printf("%lf %lf", p[1], p[2]);*/
10728:
10729:
1.136 brouard 10730: #ifdef GSL
10731: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10732: #else
1.126 brouard 10733: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10734: #endif
1.201 brouard 10735: strcpy(filerespow,"POW-MORT_");
10736: strcat(filerespow,fileresu);
1.126 brouard 10737: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10738: printf("Problem with resultfile: %s\n", filerespow);
10739: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10740: }
1.136 brouard 10741: #ifdef GSL
10742: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10743: #else
1.126 brouard 10744: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10745: #endif
1.126 brouard 10746: /* for (i=1;i<=nlstate;i++)
10747: for(j=1;j<=nlstate+ndeath;j++)
10748: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10749: */
10750: fprintf(ficrespow,"\n");
1.136 brouard 10751: #ifdef GSL
10752: /* gsl starts here */
10753: T = gsl_multimin_fminimizer_nmsimplex;
10754: gsl_multimin_fminimizer *sfm = NULL;
10755: gsl_vector *ss, *x;
10756: gsl_multimin_function minex_func;
10757:
10758: /* Initial vertex size vector */
10759: ss = gsl_vector_alloc (NDIM);
10760:
10761: if (ss == NULL){
10762: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10763: }
10764: /* Set all step sizes to 1 */
10765: gsl_vector_set_all (ss, 0.001);
10766:
10767: /* Starting point */
1.126 brouard 10768:
1.136 brouard 10769: x = gsl_vector_alloc (NDIM);
10770:
10771: if (x == NULL){
10772: gsl_vector_free(ss);
10773: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10774: }
10775:
10776: /* Initialize method and iterate */
10777: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10778: /* gsl_vector_set(x, 0, 0.0268); */
10779: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10780: gsl_vector_set(x, 0, p[1]);
10781: gsl_vector_set(x, 1, p[2]);
10782:
10783: minex_func.f = &gompertz_f;
10784: minex_func.n = NDIM;
10785: minex_func.params = (void *)&p; /* ??? */
10786:
10787: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10788: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10789:
10790: printf("Iterations beginning .....\n\n");
10791: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10792:
10793: iteri=0;
10794: while (rval == GSL_CONTINUE){
10795: iteri++;
10796: status = gsl_multimin_fminimizer_iterate(sfm);
10797:
10798: if (status) printf("error: %s\n", gsl_strerror (status));
10799: fflush(0);
10800:
10801: if (status)
10802: break;
10803:
10804: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10805: ssval = gsl_multimin_fminimizer_size (sfm);
10806:
10807: if (rval == GSL_SUCCESS)
10808: printf ("converged to a local maximum at\n");
10809:
10810: printf("%5d ", iteri);
10811: for (it = 0; it < NDIM; it++){
10812: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10813: }
10814: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10815: }
10816:
10817: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10818:
10819: gsl_vector_free(x); /* initial values */
10820: gsl_vector_free(ss); /* inital step size */
10821: for (it=0; it<NDIM; it++){
10822: p[it+1]=gsl_vector_get(sfm->x,it);
10823: fprintf(ficrespow," %.12lf", p[it]);
10824: }
10825: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10826: #endif
10827: #ifdef POWELL
10828: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10829: #endif
1.126 brouard 10830: fclose(ficrespow);
10831:
1.203 brouard 10832: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10833:
10834: for(i=1; i <=NDIM; i++)
10835: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10836: matcov[i][j]=matcov[j][i];
1.126 brouard 10837:
10838: printf("\nCovariance matrix\n ");
1.203 brouard 10839: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10840: for(i=1; i <=NDIM; i++) {
10841: for(j=1;j<=NDIM;j++){
1.220 brouard 10842: printf("%f ",matcov[i][j]);
10843: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10844: }
1.203 brouard 10845: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10846: }
10847:
10848: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10849: for (i=1;i<=NDIM;i++) {
1.126 brouard 10850: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10851: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10852: }
1.126 brouard 10853: lsurv=vector(1,AGESUP);
10854: lpop=vector(1,AGESUP);
10855: tpop=vector(1,AGESUP);
10856: lsurv[agegomp]=100000;
10857:
10858: for (k=agegomp;k<=AGESUP;k++) {
10859: agemortsup=k;
10860: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10861: }
10862:
10863: for (k=agegomp;k<agemortsup;k++)
10864: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10865:
10866: for (k=agegomp;k<agemortsup;k++){
10867: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10868: sumlpop=sumlpop+lpop[k];
10869: }
10870:
10871: tpop[agegomp]=sumlpop;
10872: for (k=agegomp;k<(agemortsup-3);k++){
10873: /* tpop[k+1]=2;*/
10874: tpop[k+1]=tpop[k]-lpop[k];
10875: }
10876:
10877:
10878: printf("\nAge lx qx dx Lx Tx e(x)\n");
10879: for (k=agegomp;k<(agemortsup-2);k++)
10880: 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]);
10881:
10882:
10883: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10884: ageminpar=50;
10885: agemaxpar=100;
1.194 brouard 10886: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10887: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10888: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10889: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10890: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10891: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10892: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10893: }else{
10894: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10895: 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 10896: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10897: }
1.201 brouard 10898: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10899: stepm, weightopt,\
10900: model,imx,p,matcov,agemortsup);
10901:
10902: free_vector(lsurv,1,AGESUP);
10903: free_vector(lpop,1,AGESUP);
10904: free_vector(tpop,1,AGESUP);
1.220 brouard 10905: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10906: free_ivector(cens,1,n);
10907: free_vector(agecens,1,n);
10908: free_ivector(dcwave,1,n);
1.220 brouard 10909: #ifdef GSL
1.136 brouard 10910: #endif
1.186 brouard 10911: } /* Endof if mle==-3 mortality only */
1.205 brouard 10912: /* Standard */
10913: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10914: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10915: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10916: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10917: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10918: for (k=1; k<=npar;k++)
10919: printf(" %d %8.5f",k,p[k]);
10920: printf("\n");
1.205 brouard 10921: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10922: /* mlikeli uses func not funcone */
1.247 brouard 10923: /* for(i=1;i<nlstate;i++){ */
10924: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10925: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10926: /* } */
1.205 brouard 10927: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10928: }
10929: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10930: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10931: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10932: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10933: }
10934: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10935: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10936: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10937: for (k=1; k<=npar;k++)
10938: printf(" %d %8.5f",k,p[k]);
10939: printf("\n");
10940:
10941: /*--------- results files --------------*/
1.224 brouard 10942: 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 10943:
10944:
10945: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10946: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10947: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10948: for(i=1,jk=1; i <=nlstate; i++){
10949: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10950: if (k != i) {
10951: printf("%d%d ",i,k);
10952: fprintf(ficlog,"%d%d ",i,k);
10953: fprintf(ficres,"%1d%1d ",i,k);
10954: for(j=1; j <=ncovmodel; j++){
10955: printf("%12.7f ",p[jk]);
10956: fprintf(ficlog,"%12.7f ",p[jk]);
10957: fprintf(ficres,"%12.7f ",p[jk]);
10958: jk++;
10959: }
10960: printf("\n");
10961: fprintf(ficlog,"\n");
10962: fprintf(ficres,"\n");
10963: }
1.126 brouard 10964: }
10965: }
1.203 brouard 10966: if(mle != 0){
10967: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10968: ftolhess=ftol; /* Usually correct */
1.203 brouard 10969: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10970: 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");
10971: 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");
10972: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10973: for(k=1; k <=(nlstate+ndeath); k++){
10974: if (k != i) {
10975: printf("%d%d ",i,k);
10976: fprintf(ficlog,"%d%d ",i,k);
10977: for(j=1; j <=ncovmodel; j++){
10978: 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]));
10979: 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]));
10980: jk++;
10981: }
10982: printf("\n");
10983: fprintf(ficlog,"\n");
10984: }
10985: }
1.193 brouard 10986: }
1.203 brouard 10987: } /* end of hesscov and Wald tests */
1.225 brouard 10988:
1.203 brouard 10989: /* */
1.126 brouard 10990: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10991: printf("# Scales (for hessian or gradient estimation)\n");
10992: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10993: for(i=1,jk=1; i <=nlstate; i++){
10994: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10995: if (j!=i) {
10996: fprintf(ficres,"%1d%1d",i,j);
10997: printf("%1d%1d",i,j);
10998: fprintf(ficlog,"%1d%1d",i,j);
10999: for(k=1; k<=ncovmodel;k++){
11000: printf(" %.5e",delti[jk]);
11001: fprintf(ficlog," %.5e",delti[jk]);
11002: fprintf(ficres," %.5e",delti[jk]);
11003: jk++;
11004: }
11005: printf("\n");
11006: fprintf(ficlog,"\n");
11007: fprintf(ficres,"\n");
11008: }
1.126 brouard 11009: }
11010: }
11011:
11012: 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 11013: if(mle >= 1) /* To big for the screen */
1.126 brouard 11014: 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");
11015: 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");
11016: /* # 121 Var(a12)\n\ */
11017: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11018: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11019: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11020: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11021: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11022: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11023: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11024:
11025:
11026: /* Just to have a covariance matrix which will be more understandable
11027: even is we still don't want to manage dictionary of variables
11028: */
11029: for(itimes=1;itimes<=2;itimes++){
11030: jj=0;
11031: for(i=1; i <=nlstate; i++){
1.225 brouard 11032: for(j=1; j <=nlstate+ndeath; j++){
11033: if(j==i) continue;
11034: for(k=1; k<=ncovmodel;k++){
11035: jj++;
11036: ca[0]= k+'a'-1;ca[1]='\0';
11037: if(itimes==1){
11038: if(mle>=1)
11039: printf("#%1d%1d%d",i,j,k);
11040: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11041: fprintf(ficres,"#%1d%1d%d",i,j,k);
11042: }else{
11043: if(mle>=1)
11044: printf("%1d%1d%d",i,j,k);
11045: fprintf(ficlog,"%1d%1d%d",i,j,k);
11046: fprintf(ficres,"%1d%1d%d",i,j,k);
11047: }
11048: ll=0;
11049: for(li=1;li <=nlstate; li++){
11050: for(lj=1;lj <=nlstate+ndeath; lj++){
11051: if(lj==li) continue;
11052: for(lk=1;lk<=ncovmodel;lk++){
11053: ll++;
11054: if(ll<=jj){
11055: cb[0]= lk +'a'-1;cb[1]='\0';
11056: if(ll<jj){
11057: if(itimes==1){
11058: if(mle>=1)
11059: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11060: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11061: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11062: }else{
11063: if(mle>=1)
11064: printf(" %.5e",matcov[jj][ll]);
11065: fprintf(ficlog," %.5e",matcov[jj][ll]);
11066: fprintf(ficres," %.5e",matcov[jj][ll]);
11067: }
11068: }else{
11069: if(itimes==1){
11070: if(mle>=1)
11071: printf(" Var(%s%1d%1d)",ca,i,j);
11072: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11073: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11074: }else{
11075: if(mle>=1)
11076: printf(" %.7e",matcov[jj][ll]);
11077: fprintf(ficlog," %.7e",matcov[jj][ll]);
11078: fprintf(ficres," %.7e",matcov[jj][ll]);
11079: }
11080: }
11081: }
11082: } /* end lk */
11083: } /* end lj */
11084: } /* end li */
11085: if(mle>=1)
11086: printf("\n");
11087: fprintf(ficlog,"\n");
11088: fprintf(ficres,"\n");
11089: numlinepar++;
11090: } /* end k*/
11091: } /*end j */
1.126 brouard 11092: } /* end i */
11093: } /* end itimes */
11094:
11095: fflush(ficlog);
11096: fflush(ficres);
1.225 brouard 11097: while(fgets(line, MAXLINE, ficpar)) {
11098: /* If line starts with a # it is a comment */
11099: if (line[0] == '#') {
11100: numlinepar++;
11101: fputs(line,stdout);
11102: fputs(line,ficparo);
11103: fputs(line,ficlog);
11104: continue;
11105: }else
11106: break;
11107: }
11108:
1.209 brouard 11109: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11110: /* ungetc(c,ficpar); */
11111: /* fgets(line, MAXLINE, ficpar); */
11112: /* fputs(line,stdout); */
11113: /* fputs(line,ficparo); */
11114: /* } */
11115: /* ungetc(c,ficpar); */
1.126 brouard 11116:
11117: estepm=0;
1.209 brouard 11118: 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 11119:
11120: if (num_filled != 6) {
11121: 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);
11122: 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);
11123: goto end;
11124: }
11125: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11126: }
11127: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11128: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11129:
1.209 brouard 11130: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11131: if (estepm==0 || estepm < stepm) estepm=stepm;
11132: if (fage <= 2) {
11133: bage = ageminpar;
11134: fage = agemaxpar;
11135: }
11136:
11137: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11138: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11139: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11140:
1.186 brouard 11141: /* Other stuffs, more or less useful */
1.254 brouard 11142: while(fgets(line, MAXLINE, ficpar)) {
11143: /* If line starts with a # it is a comment */
11144: if (line[0] == '#') {
11145: numlinepar++;
11146: fputs(line,stdout);
11147: fputs(line,ficparo);
11148: fputs(line,ficlog);
11149: continue;
11150: }else
11151: break;
11152: }
11153:
11154: 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){
11155:
11156: if (num_filled != 7) {
11157: 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);
11158: 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);
11159: goto end;
11160: }
11161: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11162: 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);
11163: 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);
11164: 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 11165: }
1.254 brouard 11166:
11167: while(fgets(line, MAXLINE, ficpar)) {
11168: /* If line starts with a # it is a comment */
11169: if (line[0] == '#') {
11170: numlinepar++;
11171: fputs(line,stdout);
11172: fputs(line,ficparo);
11173: fputs(line,ficlog);
11174: continue;
11175: }else
11176: break;
1.126 brouard 11177: }
11178:
11179:
11180: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11181: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11182:
1.254 brouard 11183: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11184: if (num_filled != 1) {
11185: 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);
11186: 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);
11187: goto end;
11188: }
11189: printf("pop_based=%d\n",popbased);
11190: fprintf(ficlog,"pop_based=%d\n",popbased);
11191: fprintf(ficparo,"pop_based=%d\n",popbased);
11192: fprintf(ficres,"pop_based=%d\n",popbased);
11193: }
11194:
1.258 brouard 11195: /* Results */
11196: nresult=0;
11197: do{
11198: if(!fgets(line, MAXLINE, ficpar)){
11199: endishere=1;
11200: parameterline=14;
11201: }else if (line[0] == '#') {
11202: /* If line starts with a # it is a comment */
1.254 brouard 11203: numlinepar++;
11204: fputs(line,stdout);
11205: fputs(line,ficparo);
11206: fputs(line,ficlog);
11207: continue;
1.258 brouard 11208: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11209: parameterline=11;
11210: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11211: parameterline=12;
11212: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11213: parameterline=13;
11214: else{
11215: parameterline=14;
1.254 brouard 11216: }
1.258 brouard 11217: switch (parameterline){
11218: case 11:
11219: 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){
11220: if (num_filled != 8) {
11221: 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);
11222: 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);
11223: goto end;
11224: }
11225: 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);
11226: 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);
11227: 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);
11228: 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);
11229: /* day and month of proj2 are not used but only year anproj2.*/
11230: }
1.254 brouard 11231: break;
1.258 brouard 11232: case 12:
11233: /*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);*/
11234: 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){
11235: if (num_filled != 8) {
11236: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11237: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11238: goto end;
11239: }
11240: 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);
11241: 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);
11242: 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);
11243: 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);
11244: /* day and month of proj2 are not used but only year anproj2.*/
11245: }
1.230 brouard 11246: break;
1.258 brouard 11247: case 13:
11248: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11249: if (num_filled == 0){
11250: resultline[0]='\0';
11251: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11252: 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);
11253: break;
11254: } else if (num_filled != 1){
11255: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11256: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11257: }
11258: nresult++; /* Sum of resultlines */
11259: printf("Result %d: result=%s\n",nresult, resultline);
11260: if(nresult > MAXRESULTLINES){
11261: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11262: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11263: goto end;
11264: }
11265: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11266: fprintf(ficparo,"result: %s\n",resultline);
11267: fprintf(ficres,"result: %s\n",resultline);
11268: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11269: break;
1.258 brouard 11270: case 14:
1.259 ! brouard 11271: if(ncovmodel >2 && nresult==0 ){
! 11272: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11273: goto end;
11274: }
1.259 ! brouard 11275: break;
1.258 brouard 11276: default:
11277: nresult=1;
11278: decoderesult(".",nresult ); /* No covariate */
11279: }
11280: } /* End switch parameterline */
11281: }while(endishere==0); /* End do */
1.126 brouard 11282:
1.230 brouard 11283: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11284: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11285:
11286: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11287: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11288: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11289: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11290: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11291: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11292: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11293: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11294: }else{
1.218 brouard 11295: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11296: }
11297: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11298: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11299: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11300:
1.225 brouard 11301: /*------------ free_vector -------------*/
11302: /* chdir(path); */
1.220 brouard 11303:
1.215 brouard 11304: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11305: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11306: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11307: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11308: free_lvector(num,1,n);
11309: free_vector(agedc,1,n);
11310: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11311: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11312: fclose(ficparo);
11313: fclose(ficres);
1.220 brouard 11314:
11315:
1.186 brouard 11316: /* Other results (useful)*/
1.220 brouard 11317:
11318:
1.126 brouard 11319: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11320: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11321: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11322: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11323: fclose(ficrespl);
11324:
11325: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11326: /*#include "hpijx.h"*/
11327: hPijx(p, bage, fage);
1.145 brouard 11328: fclose(ficrespij);
1.227 brouard 11329:
1.220 brouard 11330: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11331: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11332: k=1;
1.126 brouard 11333: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11334:
1.219 brouard 11335: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11336: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11337: for(i=1;i<=AGESUP;i++)
1.219 brouard 11338: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11339: for(k=1;k<=ncovcombmax;k++)
11340: probs[i][j][k]=0.;
1.219 brouard 11341: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11342: if (mobilav!=0 ||mobilavproj !=0 ) {
11343: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11344: for(i=1;i<=AGESUP;i++)
11345: for(j=1;j<=nlstate;j++)
11346: for(k=1;k<=ncovcombmax;k++)
11347: mobaverages[i][j][k]=0.;
1.219 brouard 11348: mobaverage=mobaverages;
11349: if (mobilav!=0) {
1.235 brouard 11350: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11351: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11352: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11353: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11354: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11355: }
1.219 brouard 11356: }
11357: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11358: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11359: else if (mobilavproj !=0) {
1.235 brouard 11360: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11361: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11362: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11363: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11364: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11365: }
1.219 brouard 11366: }
11367: }/* end if moving average */
1.227 brouard 11368:
1.126 brouard 11369: /*---------- Forecasting ------------------*/
11370: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11371: if(prevfcast==1){
11372: /* if(stepm ==1){*/
1.225 brouard 11373: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11374: }
1.217 brouard 11375: if(backcast==1){
1.219 brouard 11376: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11377: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11378: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11379:
11380: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11381:
11382: bprlim=matrix(1,nlstate,1,nlstate);
11383: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11384: fclose(ficresplb);
11385:
1.222 brouard 11386: hBijx(p, bage, fage, mobaverage);
11387: fclose(ficrespijb);
1.219 brouard 11388: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11389:
11390: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11391: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11392: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11393: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11394: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11395: }
1.217 brouard 11396:
1.186 brouard 11397:
11398: /* ------ Other prevalence ratios------------ */
1.126 brouard 11399:
1.215 brouard 11400: free_ivector(wav,1,imx);
11401: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11402: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11403: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11404:
11405:
1.127 brouard 11406: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11407:
1.201 brouard 11408: strcpy(filerese,"E_");
11409: strcat(filerese,fileresu);
1.126 brouard 11410: if((ficreseij=fopen(filerese,"w"))==NULL) {
11411: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11412: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11413: }
1.208 brouard 11414: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11415: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11416:
11417: pstamp(ficreseij);
1.219 brouard 11418:
1.235 brouard 11419: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11420: if (cptcovn < 1){i1=1;}
11421:
11422: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11423: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11424: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11425: continue;
1.219 brouard 11426: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11427: printf("\n#****** ");
1.225 brouard 11428: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11429: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11430: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11431: }
11432: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11433: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11434: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11435: }
11436: fprintf(ficreseij,"******\n");
1.235 brouard 11437: printf("******\n");
1.219 brouard 11438:
11439: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11440: oldm=oldms;savm=savms;
1.235 brouard 11441: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11442:
1.219 brouard 11443: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11444: }
11445: fclose(ficreseij);
1.208 brouard 11446: printf("done evsij\n");fflush(stdout);
11447: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11448:
1.227 brouard 11449: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11450:
11451:
1.201 brouard 11452: strcpy(filerest,"T_");
11453: strcat(filerest,fileresu);
1.127 brouard 11454: if((ficrest=fopen(filerest,"w"))==NULL) {
11455: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11456: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11457: }
1.208 brouard 11458: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11459: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11460:
1.126 brouard 11461:
1.201 brouard 11462: strcpy(fileresstde,"STDE_");
11463: strcat(fileresstde,fileresu);
1.126 brouard 11464: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11465: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11466: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11467: }
1.227 brouard 11468: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11469: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11470:
1.201 brouard 11471: strcpy(filerescve,"CVE_");
11472: strcat(filerescve,fileresu);
1.126 brouard 11473: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11474: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11475: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11476: }
1.227 brouard 11477: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11478: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11479:
1.201 brouard 11480: strcpy(fileresv,"V_");
11481: strcat(fileresv,fileresu);
1.126 brouard 11482: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11483: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11484: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11485: }
1.227 brouard 11486: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11487: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11488:
1.145 brouard 11489: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11490: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11491:
1.235 brouard 11492: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11493: if (cptcovn < 1){i1=1;}
11494:
11495: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11496: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11497: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11498: continue;
1.242 brouard 11499: printf("\n#****** Result for:");
11500: fprintf(ficrest,"\n#****** Result for:");
11501: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11502: for(j=1;j<=cptcoveff;j++){
11503: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11504: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11505: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11506: }
1.235 brouard 11507: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11508: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11509: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11510: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11511: }
1.208 brouard 11512: fprintf(ficrest,"******\n");
1.227 brouard 11513: fprintf(ficlog,"******\n");
11514: printf("******\n");
1.208 brouard 11515:
11516: fprintf(ficresstdeij,"\n#****** ");
11517: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11518: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11519: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11520: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11521: }
1.235 brouard 11522: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11523: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11524: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11525: }
1.208 brouard 11526: fprintf(ficresstdeij,"******\n");
11527: fprintf(ficrescveij,"******\n");
11528:
11529: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11530: /* pstamp(ficresvij); */
1.225 brouard 11531: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11532: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11533: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11534: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11535: }
1.208 brouard 11536: fprintf(ficresvij,"******\n");
11537:
11538: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11539: oldm=oldms;savm=savms;
1.235 brouard 11540: printf(" cvevsij ");
11541: fprintf(ficlog, " cvevsij ");
11542: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11543: printf(" end cvevsij \n ");
11544: fprintf(ficlog, " end cvevsij \n ");
11545:
11546: /*
11547: */
11548: /* goto endfree; */
11549:
11550: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11551: pstamp(ficrest);
11552:
11553:
11554: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11555: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11556: cptcod= 0; /* To be deleted */
11557: printf("varevsij vpopbased=%d \n",vpopbased);
11558: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11559: 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 11560: 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 ");
11561: if(vpopbased==1)
11562: 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);
11563: else
11564: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11565: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11566: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11567: fprintf(ficrest,"\n");
11568: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11569: epj=vector(1,nlstate+1);
11570: printf("Computing age specific period (stable) prevalences in each health state \n");
11571: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11572: for(age=bage; age <=fage ;age++){
1.235 brouard 11573: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11574: if (vpopbased==1) {
11575: if(mobilav ==0){
11576: for(i=1; i<=nlstate;i++)
11577: prlim[i][i]=probs[(int)age][i][k];
11578: }else{ /* mobilav */
11579: for(i=1; i<=nlstate;i++)
11580: prlim[i][i]=mobaverage[(int)age][i][k];
11581: }
11582: }
1.219 brouard 11583:
1.227 brouard 11584: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11585: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11586: /* printf(" age %4.0f ",age); */
11587: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11588: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11589: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11590: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11591: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11592: }
11593: epj[nlstate+1] +=epj[j];
11594: }
11595: /* printf(" age %4.0f \n",age); */
1.219 brouard 11596:
1.227 brouard 11597: for(i=1, vepp=0.;i <=nlstate;i++)
11598: for(j=1;j <=nlstate;j++)
11599: vepp += vareij[i][j][(int)age];
11600: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11601: for(j=1;j <=nlstate;j++){
11602: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11603: }
11604: fprintf(ficrest,"\n");
11605: }
1.208 brouard 11606: } /* End vpopbased */
11607: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11608: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11609: free_vector(epj,1,nlstate+1);
1.235 brouard 11610: printf("done selection\n");fflush(stdout);
11611: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11612:
1.145 brouard 11613: /*}*/
1.235 brouard 11614: } /* End k selection */
1.227 brouard 11615:
11616: printf("done State-specific expectancies\n");fflush(stdout);
11617: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11618:
1.126 brouard 11619: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11620:
1.201 brouard 11621: strcpy(fileresvpl,"VPL_");
11622: strcat(fileresvpl,fileresu);
1.126 brouard 11623: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11624: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11625: exit(0);
11626: }
1.208 brouard 11627: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11628: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11629:
1.145 brouard 11630: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11631: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11632:
1.235 brouard 11633: i1=pow(2,cptcoveff);
11634: if (cptcovn < 1){i1=1;}
11635:
11636: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11637: for(k=1; k<=i1;k++){
1.253 brouard 11638: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11639: continue;
1.227 brouard 11640: fprintf(ficresvpl,"\n#****** ");
11641: printf("\n#****** ");
11642: fprintf(ficlog,"\n#****** ");
11643: for(j=1;j<=cptcoveff;j++) {
11644: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11645: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11646: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11647: }
1.235 brouard 11648: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11649: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11650: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11651: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11652: }
1.227 brouard 11653: fprintf(ficresvpl,"******\n");
11654: printf("******\n");
11655: fprintf(ficlog,"******\n");
11656:
11657: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11658: oldm=oldms;savm=savms;
1.235 brouard 11659: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11660: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11661: /*}*/
1.126 brouard 11662: }
1.227 brouard 11663:
1.126 brouard 11664: fclose(ficresvpl);
1.208 brouard 11665: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11666: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11667:
11668: free_vector(weight,1,n);
11669: free_imatrix(Tvard,1,NCOVMAX,1,2);
11670: free_imatrix(s,1,maxwav+1,1,n);
11671: free_matrix(anint,1,maxwav,1,n);
11672: free_matrix(mint,1,maxwav,1,n);
11673: free_ivector(cod,1,n);
11674: free_ivector(tab,1,NCOVMAX);
11675: fclose(ficresstdeij);
11676: fclose(ficrescveij);
11677: fclose(ficresvij);
11678: fclose(ficrest);
11679: fclose(ficpar);
11680:
11681:
1.126 brouard 11682: /*---------- End : free ----------------*/
1.219 brouard 11683: if (mobilav!=0 ||mobilavproj !=0)
11684: 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 11685: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11686: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11687: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11688: } /* mle==-3 arrives here for freeing */
1.227 brouard 11689: /* endfree:*/
11690: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11691: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11692: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11693: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11694: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11695: free_matrix(coqvar,1,maxwav,1,n);
11696: free_matrix(covar,0,NCOVMAX,1,n);
11697: free_matrix(matcov,1,npar,1,npar);
11698: free_matrix(hess,1,npar,1,npar);
11699: /*free_vector(delti,1,npar);*/
11700: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11701: free_matrix(agev,1,maxwav,1,imx);
11702: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11703:
11704: free_ivector(ncodemax,1,NCOVMAX);
11705: free_ivector(ncodemaxwundef,1,NCOVMAX);
11706: free_ivector(Dummy,-1,NCOVMAX);
11707: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11708: free_ivector(DummyV,1,NCOVMAX);
11709: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11710: free_ivector(Typevar,-1,NCOVMAX);
11711: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11712: free_ivector(TvarsQ,1,NCOVMAX);
11713: free_ivector(TvarsQind,1,NCOVMAX);
11714: free_ivector(TvarsD,1,NCOVMAX);
11715: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11716: free_ivector(TvarFD,1,NCOVMAX);
11717: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11718: free_ivector(TvarF,1,NCOVMAX);
11719: free_ivector(TvarFind,1,NCOVMAX);
11720: free_ivector(TvarV,1,NCOVMAX);
11721: free_ivector(TvarVind,1,NCOVMAX);
11722: free_ivector(TvarA,1,NCOVMAX);
11723: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11724: free_ivector(TvarFQ,1,NCOVMAX);
11725: free_ivector(TvarFQind,1,NCOVMAX);
11726: free_ivector(TvarVD,1,NCOVMAX);
11727: free_ivector(TvarVDind,1,NCOVMAX);
11728: free_ivector(TvarVQ,1,NCOVMAX);
11729: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11730: free_ivector(Tvarsel,1,NCOVMAX);
11731: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11732: free_ivector(Tposprod,1,NCOVMAX);
11733: free_ivector(Tprod,1,NCOVMAX);
11734: free_ivector(Tvaraff,1,NCOVMAX);
11735: free_ivector(invalidvarcomb,1,ncovcombmax);
11736: free_ivector(Tage,1,NCOVMAX);
11737: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11738: free_ivector(TmodelInvind,1,NCOVMAX);
11739: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11740:
11741: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11742: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11743: fflush(fichtm);
11744: fflush(ficgp);
11745:
1.227 brouard 11746:
1.126 brouard 11747: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11748: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11749: 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 11750: }else{
11751: printf("End of Imach\n");
11752: fprintf(ficlog,"End of Imach\n");
11753: }
11754: printf("See log file on %s\n",filelog);
11755: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11756: /*(void) gettimeofday(&end_time,&tzp);*/
11757: rend_time = time(NULL);
11758: end_time = *localtime(&rend_time);
11759: /* tml = *localtime(&end_time.tm_sec); */
11760: strcpy(strtend,asctime(&end_time));
1.126 brouard 11761: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11762: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11763: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11764:
1.157 brouard 11765: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11766: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11767: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11768: /* printf("Total time was %d uSec.\n", total_usecs);*/
11769: /* if(fileappend(fichtm,optionfilehtm)){ */
11770: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11771: fclose(fichtm);
11772: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11773: fclose(fichtmcov);
11774: fclose(ficgp);
11775: fclose(ficlog);
11776: /*------ End -----------*/
1.227 brouard 11777:
11778:
11779: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11780: #ifdef WIN32
1.227 brouard 11781: if (_chdir(pathcd) != 0)
11782: printf("Can't move to directory %s!\n",path);
11783: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11784: #else
1.227 brouard 11785: if(chdir(pathcd) != 0)
11786: printf("Can't move to directory %s!\n", path);
11787: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11788: #endif
1.126 brouard 11789: printf("Current directory %s!\n",pathcd);
11790: /*strcat(plotcmd,CHARSEPARATOR);*/
11791: sprintf(plotcmd,"gnuplot");
1.157 brouard 11792: #ifdef _WIN32
1.126 brouard 11793: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11794: #endif
11795: if(!stat(plotcmd,&info)){
1.158 brouard 11796: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11797: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11798: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11799: }else
11800: strcpy(pplotcmd,plotcmd);
1.157 brouard 11801: #ifdef __unix
1.126 brouard 11802: strcpy(plotcmd,GNUPLOTPROGRAM);
11803: if(!stat(plotcmd,&info)){
1.158 brouard 11804: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11805: }else
11806: strcpy(pplotcmd,plotcmd);
11807: #endif
11808: }else
11809: strcpy(pplotcmd,plotcmd);
11810:
11811: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11812: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11813:
1.126 brouard 11814: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11815: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11816: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11817: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11818: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11819: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11820: }
1.158 brouard 11821: printf(" Successful, please wait...");
1.126 brouard 11822: while (z[0] != 'q') {
11823: /* chdir(path); */
1.154 brouard 11824: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11825: scanf("%s",z);
11826: /* if (z[0] == 'c') system("./imach"); */
11827: if (z[0] == 'e') {
1.158 brouard 11828: #ifdef __APPLE__
1.152 brouard 11829: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11830: #elif __linux
11831: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11832: #else
1.152 brouard 11833: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11834: #endif
11835: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11836: system(pplotcmd);
1.126 brouard 11837: }
11838: else if (z[0] == 'g') system(plotcmd);
11839: else if (z[0] == 'q') exit(0);
11840: }
1.227 brouard 11841: end:
1.126 brouard 11842: while (z[0] != 'q') {
1.195 brouard 11843: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11844: scanf("%s",z);
11845: }
11846: }
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