Annotation of imach/src/imach.c, revision 1.273
1.273 ! brouard 1: /* $Id: imach.c,v 1.272 2017/06/27 10:22:40 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.273 ! brouard 4: Revision 1.272 2017/06/27 10:22:40 brouard
! 5: Summary: Color of backprojection changed from 6 to 5(yellow)
! 6:
1.272 brouard 7: Revision 1.271 2017/06/27 10:17:50 brouard
8: Summary: Some bug with rint
9:
1.271 brouard 10: Revision 1.270 2017/05/24 05:45:29 brouard
11: *** empty log message ***
12:
1.270 brouard 13: Revision 1.269 2017/05/23 08:39:25 brouard
14: Summary: Code into subroutine, cleanings
15:
1.269 brouard 16: Revision 1.268 2017/05/18 20:09:32 brouard
17: Summary: backprojection and confidence intervals of backprevalence
18:
1.268 brouard 19: Revision 1.267 2017/05/13 10:25:05 brouard
20: Summary: temporary save for backprojection
21:
1.267 brouard 22: Revision 1.266 2017/05/13 07:26:12 brouard
23: Summary: Version 0.99r13 (improvements and bugs fixed)
24:
1.266 brouard 25: Revision 1.265 2017/04/26 16:22:11 brouard
26: Summary: imach 0.99r13 Some bugs fixed
27:
1.265 brouard 28: Revision 1.264 2017/04/26 06:01:29 brouard
29: Summary: Labels in graphs
30:
1.264 brouard 31: Revision 1.263 2017/04/24 15:23:15 brouard
32: Summary: to save
33:
1.263 brouard 34: Revision 1.262 2017/04/18 16:48:12 brouard
35: *** empty log message ***
36:
1.262 brouard 37: Revision 1.261 2017/04/05 10:14:09 brouard
38: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
39:
1.261 brouard 40: Revision 1.260 2017/04/04 17:46:59 brouard
41: Summary: Gnuplot indexations fixed (humm)
42:
1.260 brouard 43: Revision 1.259 2017/04/04 13:01:16 brouard
44: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
45:
1.259 brouard 46: Revision 1.258 2017/04/03 10:17:47 brouard
47: Summary: Version 0.99r12
48:
49: Some cleanings, conformed with updated documentation.
50:
1.258 brouard 51: Revision 1.257 2017/03/29 16:53:30 brouard
52: Summary: Temp
53:
1.257 brouard 54: Revision 1.256 2017/03/27 05:50:23 brouard
55: Summary: Temporary
56:
1.256 brouard 57: Revision 1.255 2017/03/08 16:02:28 brouard
58: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
59:
1.255 brouard 60: Revision 1.254 2017/03/08 07:13:00 brouard
61: Summary: Fixing data parameter line
62:
1.254 brouard 63: Revision 1.253 2016/12/15 11:59:41 brouard
64: Summary: 0.99 in progress
65:
1.253 brouard 66: Revision 1.252 2016/09/15 21:15:37 brouard
67: *** empty log message ***
68:
1.252 brouard 69: Revision 1.251 2016/09/15 15:01:13 brouard
70: Summary: not working
71:
1.251 brouard 72: Revision 1.250 2016/09/08 16:07:27 brouard
73: Summary: continue
74:
1.250 brouard 75: Revision 1.249 2016/09/07 17:14:18 brouard
76: Summary: Starting values from frequencies
77:
1.249 brouard 78: Revision 1.248 2016/09/07 14:10:18 brouard
79: *** empty log message ***
80:
1.248 brouard 81: Revision 1.247 2016/09/02 11:11:21 brouard
82: *** empty log message ***
83:
1.247 brouard 84: Revision 1.246 2016/09/02 08:49:22 brouard
85: *** empty log message ***
86:
1.246 brouard 87: Revision 1.245 2016/09/02 07:25:01 brouard
88: *** empty log message ***
89:
1.245 brouard 90: Revision 1.244 2016/09/02 07:17:34 brouard
91: *** empty log message ***
92:
1.244 brouard 93: Revision 1.243 2016/09/02 06:45:35 brouard
94: *** empty log message ***
95:
1.243 brouard 96: Revision 1.242 2016/08/30 15:01:20 brouard
97: Summary: Fixing a lots
98:
1.242 brouard 99: Revision 1.241 2016/08/29 17:17:25 brouard
100: Summary: gnuplot problem in Back projection to fix
101:
1.241 brouard 102: Revision 1.240 2016/08/29 07:53:18 brouard
103: Summary: Better
104:
1.240 brouard 105: Revision 1.239 2016/08/26 15:51:03 brouard
106: Summary: Improvement in Powell output in order to copy and paste
107:
108: Author:
109:
1.239 brouard 110: Revision 1.238 2016/08/26 14:23:35 brouard
111: Summary: Starting tests of 0.99
112:
1.238 brouard 113: Revision 1.237 2016/08/26 09:20:19 brouard
114: Summary: to valgrind
115:
1.237 brouard 116: Revision 1.236 2016/08/25 10:50:18 brouard
117: *** empty log message ***
118:
1.236 brouard 119: Revision 1.235 2016/08/25 06:59:23 brouard
120: *** empty log message ***
121:
1.235 brouard 122: Revision 1.234 2016/08/23 16:51:20 brouard
123: *** empty log message ***
124:
1.234 brouard 125: Revision 1.233 2016/08/23 07:40:50 brouard
126: Summary: not working
127:
1.233 brouard 128: Revision 1.232 2016/08/22 14:20:21 brouard
129: Summary: not working
130:
1.232 brouard 131: Revision 1.231 2016/08/22 07:17:15 brouard
132: Summary: not working
133:
1.231 brouard 134: Revision 1.230 2016/08/22 06:55:53 brouard
135: Summary: Not working
136:
1.230 brouard 137: Revision 1.229 2016/07/23 09:45:53 brouard
138: Summary: Completing for func too
139:
1.229 brouard 140: Revision 1.228 2016/07/22 17:45:30 brouard
141: Summary: Fixing some arrays, still debugging
142:
1.227 brouard 143: Revision 1.226 2016/07/12 18:42:34 brouard
144: Summary: temp
145:
1.226 brouard 146: Revision 1.225 2016/07/12 08:40:03 brouard
147: Summary: saving but not running
148:
1.225 brouard 149: Revision 1.224 2016/07/01 13:16:01 brouard
150: Summary: Fixes
151:
1.224 brouard 152: Revision 1.223 2016/02/19 09:23:35 brouard
153: Summary: temporary
154:
1.223 brouard 155: Revision 1.222 2016/02/17 08:14:50 brouard
156: Summary: Probably last 0.98 stable version 0.98r6
157:
1.222 brouard 158: Revision 1.221 2016/02/15 23:35:36 brouard
159: Summary: minor bug
160:
1.220 brouard 161: Revision 1.219 2016/02/15 00:48:12 brouard
162: *** empty log message ***
163:
1.219 brouard 164: Revision 1.218 2016/02/12 11:29:23 brouard
165: Summary: 0.99 Back projections
166:
1.218 brouard 167: Revision 1.217 2015/12/23 17:18:31 brouard
168: Summary: Experimental backcast
169:
1.217 brouard 170: Revision 1.216 2015/12/18 17:32:11 brouard
171: Summary: 0.98r4 Warning and status=-2
172:
173: Version 0.98r4 is now:
174: - displaying an error when status is -1, date of interview unknown and date of death known;
175: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
176: Older changes concerning s=-2, dating from 2005 have been supersed.
177:
1.216 brouard 178: Revision 1.215 2015/12/16 08:52:24 brouard
179: Summary: 0.98r4 working
180:
1.215 brouard 181: Revision 1.214 2015/12/16 06:57:54 brouard
182: Summary: temporary not working
183:
1.214 brouard 184: Revision 1.213 2015/12/11 18:22:17 brouard
185: Summary: 0.98r4
186:
1.213 brouard 187: Revision 1.212 2015/11/21 12:47:24 brouard
188: Summary: minor typo
189:
1.212 brouard 190: Revision 1.211 2015/11/21 12:41:11 brouard
191: Summary: 0.98r3 with some graph of projected cross-sectional
192:
193: Author: Nicolas Brouard
194:
1.211 brouard 195: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 196: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 197: Summary: Adding ftolpl parameter
198: Author: N Brouard
199:
200: We had difficulties to get smoothed confidence intervals. It was due
201: to the period prevalence which wasn't computed accurately. The inner
202: parameter ftolpl is now an outer parameter of the .imach parameter
203: file after estepm. If ftolpl is small 1.e-4 and estepm too,
204: computation are long.
205:
1.209 brouard 206: Revision 1.208 2015/11/17 14:31:57 brouard
207: Summary: temporary
208:
1.208 brouard 209: Revision 1.207 2015/10/27 17:36:57 brouard
210: *** empty log message ***
211:
1.207 brouard 212: Revision 1.206 2015/10/24 07:14:11 brouard
213: *** empty log message ***
214:
1.206 brouard 215: Revision 1.205 2015/10/23 15:50:53 brouard
216: Summary: 0.98r3 some clarification for graphs on likelihood contributions
217:
1.205 brouard 218: Revision 1.204 2015/10/01 16:20:26 brouard
219: Summary: Some new graphs of contribution to likelihood
220:
1.204 brouard 221: Revision 1.203 2015/09/30 17:45:14 brouard
222: Summary: looking at better estimation of the hessian
223:
224: Also a better criteria for convergence to the period prevalence And
225: therefore adding the number of years needed to converge. (The
226: prevalence in any alive state shold sum to one
227:
1.203 brouard 228: Revision 1.202 2015/09/22 19:45:16 brouard
229: Summary: Adding some overall graph on contribution to likelihood. Might change
230:
1.202 brouard 231: Revision 1.201 2015/09/15 17:34:58 brouard
232: Summary: 0.98r0
233:
234: - Some new graphs like suvival functions
235: - Some bugs fixed like model=1+age+V2.
236:
1.201 brouard 237: Revision 1.200 2015/09/09 16:53:55 brouard
238: Summary: Big bug thanks to Flavia
239:
240: Even model=1+age+V2. did not work anymore
241:
1.200 brouard 242: Revision 1.199 2015/09/07 14:09:23 brouard
243: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
244:
1.199 brouard 245: Revision 1.198 2015/09/03 07:14:39 brouard
246: Summary: 0.98q5 Flavia
247:
1.198 brouard 248: Revision 1.197 2015/09/01 18:24:39 brouard
249: *** empty log message ***
250:
1.197 brouard 251: Revision 1.196 2015/08/18 23:17:52 brouard
252: Summary: 0.98q5
253:
1.196 brouard 254: Revision 1.195 2015/08/18 16:28:39 brouard
255: Summary: Adding a hack for testing purpose
256:
257: After reading the title, ftol and model lines, if the comment line has
258: a q, starting with #q, the answer at the end of the run is quit. It
259: permits to run test files in batch with ctest. The former workaround was
260: $ echo q | imach foo.imach
261:
1.195 brouard 262: Revision 1.194 2015/08/18 13:32:00 brouard
263: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
264:
1.194 brouard 265: Revision 1.193 2015/08/04 07:17:42 brouard
266: Summary: 0.98q4
267:
1.193 brouard 268: Revision 1.192 2015/07/16 16:49:02 brouard
269: Summary: Fixing some outputs
270:
1.192 brouard 271: Revision 1.191 2015/07/14 10:00:33 brouard
272: Summary: Some fixes
273:
1.191 brouard 274: Revision 1.190 2015/05/05 08:51:13 brouard
275: Summary: Adding digits in output parameters (7 digits instead of 6)
276:
277: Fix 1+age+.
278:
1.190 brouard 279: Revision 1.189 2015/04/30 14:45:16 brouard
280: Summary: 0.98q2
281:
1.189 brouard 282: Revision 1.188 2015/04/30 08:27:53 brouard
283: *** empty log message ***
284:
1.188 brouard 285: Revision 1.187 2015/04/29 09:11:15 brouard
286: *** empty log message ***
287:
1.187 brouard 288: Revision 1.186 2015/04/23 12:01:52 brouard
289: Summary: V1*age is working now, version 0.98q1
290:
291: Some codes had been disabled in order to simplify and Vn*age was
292: working in the optimization phase, ie, giving correct MLE parameters,
293: but, as usual, outputs were not correct and program core dumped.
294:
1.186 brouard 295: Revision 1.185 2015/03/11 13:26:42 brouard
296: Summary: Inclusion of compile and links command line for Intel Compiler
297:
1.185 brouard 298: Revision 1.184 2015/03/11 11:52:39 brouard
299: Summary: Back from Windows 8. Intel Compiler
300:
1.184 brouard 301: Revision 1.183 2015/03/10 20:34:32 brouard
302: Summary: 0.98q0, trying with directest, mnbrak fixed
303:
304: We use directest instead of original Powell test; probably no
305: incidence on the results, but better justifications;
306: We fixed Numerical Recipes mnbrak routine which was wrong and gave
307: wrong results.
308:
1.183 brouard 309: Revision 1.182 2015/02/12 08:19:57 brouard
310: Summary: Trying to keep directest which seems simpler and more general
311: Author: Nicolas Brouard
312:
1.182 brouard 313: Revision 1.181 2015/02/11 23:22:24 brouard
314: Summary: Comments on Powell added
315:
316: Author:
317:
1.181 brouard 318: Revision 1.180 2015/02/11 17:33:45 brouard
319: Summary: Finishing move from main to function (hpijx and prevalence_limit)
320:
1.180 brouard 321: Revision 1.179 2015/01/04 09:57:06 brouard
322: Summary: back to OS/X
323:
1.179 brouard 324: Revision 1.178 2015/01/04 09:35:48 brouard
325: *** empty log message ***
326:
1.178 brouard 327: Revision 1.177 2015/01/03 18:40:56 brouard
328: Summary: Still testing ilc32 on OSX
329:
1.177 brouard 330: Revision 1.176 2015/01/03 16:45:04 brouard
331: *** empty log message ***
332:
1.176 brouard 333: Revision 1.175 2015/01/03 16:33:42 brouard
334: *** empty log message ***
335:
1.175 brouard 336: Revision 1.174 2015/01/03 16:15:49 brouard
337: Summary: Still in cross-compilation
338:
1.174 brouard 339: Revision 1.173 2015/01/03 12:06:26 brouard
340: Summary: trying to detect cross-compilation
341:
1.173 brouard 342: Revision 1.172 2014/12/27 12:07:47 brouard
343: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
344:
1.172 brouard 345: Revision 1.171 2014/12/23 13:26:59 brouard
346: Summary: Back from Visual C
347:
348: Still problem with utsname.h on Windows
349:
1.171 brouard 350: Revision 1.170 2014/12/23 11:17:12 brouard
351: Summary: Cleaning some \%% back to %%
352:
353: The escape was mandatory for a specific compiler (which one?), but too many warnings.
354:
1.170 brouard 355: Revision 1.169 2014/12/22 23:08:31 brouard
356: Summary: 0.98p
357:
358: Outputs some informations on compiler used, OS etc. Testing on different platforms.
359:
1.169 brouard 360: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 361: Summary: update
1.169 brouard 362:
1.168 brouard 363: Revision 1.167 2014/12/22 13:50:56 brouard
364: Summary: Testing uname and compiler version and if compiled 32 or 64
365:
366: Testing on Linux 64
367:
1.167 brouard 368: Revision 1.166 2014/12/22 11:40:47 brouard
369: *** empty log message ***
370:
1.166 brouard 371: Revision 1.165 2014/12/16 11:20:36 brouard
372: Summary: After compiling on Visual C
373:
374: * imach.c (Module): Merging 1.61 to 1.162
375:
1.165 brouard 376: Revision 1.164 2014/12/16 10:52:11 brouard
377: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
378:
379: * imach.c (Module): Merging 1.61 to 1.162
380:
1.164 brouard 381: Revision 1.163 2014/12/16 10:30:11 brouard
382: * imach.c (Module): Merging 1.61 to 1.162
383:
1.163 brouard 384: Revision 1.162 2014/09/25 11:43:39 brouard
385: Summary: temporary backup 0.99!
386:
1.162 brouard 387: Revision 1.1 2014/09/16 11:06:58 brouard
388: Summary: With some code (wrong) for nlopt
389:
390: Author:
391:
392: Revision 1.161 2014/09/15 20:41:41 brouard
393: Summary: Problem with macro SQR on Intel compiler
394:
1.161 brouard 395: Revision 1.160 2014/09/02 09:24:05 brouard
396: *** empty log message ***
397:
1.160 brouard 398: Revision 1.159 2014/09/01 10:34:10 brouard
399: Summary: WIN32
400: Author: Brouard
401:
1.159 brouard 402: Revision 1.158 2014/08/27 17:11:51 brouard
403: *** empty log message ***
404:
1.158 brouard 405: Revision 1.157 2014/08/27 16:26:55 brouard
406: Summary: Preparing windows Visual studio version
407: Author: Brouard
408:
409: In order to compile on Visual studio, time.h is now correct and time_t
410: and tm struct should be used. difftime should be used but sometimes I
411: just make the differences in raw time format (time(&now).
412: Trying to suppress #ifdef LINUX
413: Add xdg-open for __linux in order to open default browser.
414:
1.157 brouard 415: Revision 1.156 2014/08/25 20:10:10 brouard
416: *** empty log message ***
417:
1.156 brouard 418: Revision 1.155 2014/08/25 18:32:34 brouard
419: Summary: New compile, minor changes
420: Author: Brouard
421:
1.155 brouard 422: Revision 1.154 2014/06/20 17:32:08 brouard
423: Summary: Outputs now all graphs of convergence to period prevalence
424:
1.154 brouard 425: Revision 1.153 2014/06/20 16:45:46 brouard
426: Summary: If 3 live state, convergence to period prevalence on same graph
427: Author: Brouard
428:
1.153 brouard 429: Revision 1.152 2014/06/18 17:54:09 brouard
430: Summary: open browser, use gnuplot on same dir than imach if not found in the path
431:
1.152 brouard 432: Revision 1.151 2014/06/18 16:43:30 brouard
433: *** empty log message ***
434:
1.151 brouard 435: Revision 1.150 2014/06/18 16:42:35 brouard
436: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
437: Author: brouard
438:
1.150 brouard 439: Revision 1.149 2014/06/18 15:51:14 brouard
440: Summary: Some fixes in parameter files errors
441: Author: Nicolas Brouard
442:
1.149 brouard 443: Revision 1.148 2014/06/17 17:38:48 brouard
444: Summary: Nothing new
445: Author: Brouard
446:
447: Just a new packaging for OS/X version 0.98nS
448:
1.148 brouard 449: Revision 1.147 2014/06/16 10:33:11 brouard
450: *** empty log message ***
451:
1.147 brouard 452: Revision 1.146 2014/06/16 10:20:28 brouard
453: Summary: Merge
454: Author: Brouard
455:
456: Merge, before building revised version.
457:
1.146 brouard 458: Revision 1.145 2014/06/10 21:23:15 brouard
459: Summary: Debugging with valgrind
460: Author: Nicolas Brouard
461:
462: Lot of changes in order to output the results with some covariates
463: After the Edimburgh REVES conference 2014, it seems mandatory to
464: improve the code.
465: No more memory valgrind error but a lot has to be done in order to
466: continue the work of splitting the code into subroutines.
467: Also, decodemodel has been improved. Tricode is still not
468: optimal. nbcode should be improved. Documentation has been added in
469: the source code.
470:
1.144 brouard 471: Revision 1.143 2014/01/26 09:45:38 brouard
472: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
473:
474: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
475: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
476:
1.143 brouard 477: Revision 1.142 2014/01/26 03:57:36 brouard
478: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
479:
480: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
481:
1.142 brouard 482: Revision 1.141 2014/01/26 02:42:01 brouard
483: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
484:
1.141 brouard 485: Revision 1.140 2011/09/02 10:37:54 brouard
486: Summary: times.h is ok with mingw32 now.
487:
1.140 brouard 488: Revision 1.139 2010/06/14 07:50:17 brouard
489: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
490: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
491:
1.139 brouard 492: Revision 1.138 2010/04/30 18:19:40 brouard
493: *** empty log message ***
494:
1.138 brouard 495: Revision 1.137 2010/04/29 18:11:38 brouard
496: (Module): Checking covariates for more complex models
497: than V1+V2. A lot of change to be done. Unstable.
498:
1.137 brouard 499: Revision 1.136 2010/04/26 20:30:53 brouard
500: (Module): merging some libgsl code. Fixing computation
501: of likelione (using inter/intrapolation if mle = 0) in order to
502: get same likelihood as if mle=1.
503: Some cleaning of code and comments added.
504:
1.136 brouard 505: Revision 1.135 2009/10/29 15:33:14 brouard
506: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
507:
1.135 brouard 508: Revision 1.134 2009/10/29 13:18:53 brouard
509: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
510:
1.134 brouard 511: Revision 1.133 2009/07/06 10:21:25 brouard
512: just nforces
513:
1.133 brouard 514: Revision 1.132 2009/07/06 08:22:05 brouard
515: Many tings
516:
1.132 brouard 517: Revision 1.131 2009/06/20 16:22:47 brouard
518: Some dimensions resccaled
519:
1.131 brouard 520: Revision 1.130 2009/05/26 06:44:34 brouard
521: (Module): Max Covariate is now set to 20 instead of 8. A
522: lot of cleaning with variables initialized to 0. Trying to make
523: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
524:
1.130 brouard 525: Revision 1.129 2007/08/31 13:49:27 lievre
526: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
527:
1.129 lievre 528: Revision 1.128 2006/06/30 13:02:05 brouard
529: (Module): Clarifications on computing e.j
530:
1.128 brouard 531: Revision 1.127 2006/04/28 18:11:50 brouard
532: (Module): Yes the sum of survivors was wrong since
533: imach-114 because nhstepm was no more computed in the age
534: loop. Now we define nhstepma in the age loop.
535: (Module): In order to speed up (in case of numerous covariates) we
536: compute health expectancies (without variances) in a first step
537: and then all the health expectancies with variances or standard
538: deviation (needs data from the Hessian matrices) which slows the
539: computation.
540: In the future we should be able to stop the program is only health
541: expectancies and graph are needed without standard deviations.
542:
1.127 brouard 543: Revision 1.126 2006/04/28 17:23:28 brouard
544: (Module): Yes the sum of survivors was wrong since
545: imach-114 because nhstepm was no more computed in the age
546: loop. Now we define nhstepma in the age loop.
547: Version 0.98h
548:
1.126 brouard 549: Revision 1.125 2006/04/04 15:20:31 lievre
550: Errors in calculation of health expectancies. Age was not initialized.
551: Forecasting file added.
552:
553: Revision 1.124 2006/03/22 17:13:53 lievre
554: Parameters are printed with %lf instead of %f (more numbers after the comma).
555: The log-likelihood is printed in the log file
556:
557: Revision 1.123 2006/03/20 10:52:43 brouard
558: * imach.c (Module): <title> changed, corresponds to .htm file
559: name. <head> headers where missing.
560:
561: * imach.c (Module): Weights can have a decimal point as for
562: English (a comma might work with a correct LC_NUMERIC environment,
563: otherwise the weight is truncated).
564: Modification of warning when the covariates values are not 0 or
565: 1.
566: Version 0.98g
567:
568: Revision 1.122 2006/03/20 09:45:41 brouard
569: (Module): Weights can have a decimal point as for
570: English (a comma might work with a correct LC_NUMERIC environment,
571: otherwise the weight is truncated).
572: Modification of warning when the covariates values are not 0 or
573: 1.
574: Version 0.98g
575:
576: Revision 1.121 2006/03/16 17:45:01 lievre
577: * imach.c (Module): Comments concerning covariates added
578:
579: * imach.c (Module): refinements in the computation of lli if
580: status=-2 in order to have more reliable computation if stepm is
581: not 1 month. Version 0.98f
582:
583: Revision 1.120 2006/03/16 15:10:38 lievre
584: (Module): refinements in the computation of lli if
585: status=-2 in order to have more reliable computation if stepm is
586: not 1 month. Version 0.98f
587:
588: Revision 1.119 2006/03/15 17:42:26 brouard
589: (Module): Bug if status = -2, the loglikelihood was
590: computed as likelihood omitting the logarithm. Version O.98e
591:
592: Revision 1.118 2006/03/14 18:20:07 brouard
593: (Module): varevsij Comments added explaining the second
594: table of variances if popbased=1 .
595: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
596: (Module): Function pstamp added
597: (Module): Version 0.98d
598:
599: Revision 1.117 2006/03/14 17:16:22 brouard
600: (Module): varevsij Comments added explaining the second
601: table of variances if popbased=1 .
602: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
603: (Module): Function pstamp added
604: (Module): Version 0.98d
605:
606: Revision 1.116 2006/03/06 10:29:27 brouard
607: (Module): Variance-covariance wrong links and
608: varian-covariance of ej. is needed (Saito).
609:
610: Revision 1.115 2006/02/27 12:17:45 brouard
611: (Module): One freematrix added in mlikeli! 0.98c
612:
613: Revision 1.114 2006/02/26 12:57:58 brouard
614: (Module): Some improvements in processing parameter
615: filename with strsep.
616:
617: Revision 1.113 2006/02/24 14:20:24 brouard
618: (Module): Memory leaks checks with valgrind and:
619: datafile was not closed, some imatrix were not freed and on matrix
620: allocation too.
621:
622: Revision 1.112 2006/01/30 09:55:26 brouard
623: (Module): Back to gnuplot.exe instead of wgnuplot.exe
624:
625: Revision 1.111 2006/01/25 20:38:18 brouard
626: (Module): Lots of cleaning and bugs added (Gompertz)
627: (Module): Comments can be added in data file. Missing date values
628: can be a simple dot '.'.
629:
630: Revision 1.110 2006/01/25 00:51:50 brouard
631: (Module): Lots of cleaning and bugs added (Gompertz)
632:
633: Revision 1.109 2006/01/24 19:37:15 brouard
634: (Module): Comments (lines starting with a #) are allowed in data.
635:
636: Revision 1.108 2006/01/19 18:05:42 lievre
637: Gnuplot problem appeared...
638: To be fixed
639:
640: Revision 1.107 2006/01/19 16:20:37 brouard
641: Test existence of gnuplot in imach path
642:
643: Revision 1.106 2006/01/19 13:24:36 brouard
644: Some cleaning and links added in html output
645:
646: Revision 1.105 2006/01/05 20:23:19 lievre
647: *** empty log message ***
648:
649: Revision 1.104 2005/09/30 16:11:43 lievre
650: (Module): sump fixed, loop imx fixed, and simplifications.
651: (Module): If the status is missing at the last wave but we know
652: that the person is alive, then we can code his/her status as -2
653: (instead of missing=-1 in earlier versions) and his/her
654: contributions to the likelihood is 1 - Prob of dying from last
655: health status (= 1-p13= p11+p12 in the easiest case of somebody in
656: the healthy state at last known wave). Version is 0.98
657:
658: Revision 1.103 2005/09/30 15:54:49 lievre
659: (Module): sump fixed, loop imx fixed, and simplifications.
660:
661: Revision 1.102 2004/09/15 17:31:30 brouard
662: Add the possibility to read data file including tab characters.
663:
664: Revision 1.101 2004/09/15 10:38:38 brouard
665: Fix on curr_time
666:
667: Revision 1.100 2004/07/12 18:29:06 brouard
668: Add version for Mac OS X. Just define UNIX in Makefile
669:
670: Revision 1.99 2004/06/05 08:57:40 brouard
671: *** empty log message ***
672:
673: Revision 1.98 2004/05/16 15:05:56 brouard
674: New version 0.97 . First attempt to estimate force of mortality
675: directly from the data i.e. without the need of knowing the health
676: state at each age, but using a Gompertz model: log u =a + b*age .
677: This is the basic analysis of mortality and should be done before any
678: other analysis, in order to test if the mortality estimated from the
679: cross-longitudinal survey is different from the mortality estimated
680: from other sources like vital statistic data.
681:
682: The same imach parameter file can be used but the option for mle should be -3.
683:
1.133 brouard 684: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 685: former routines in order to include the new code within the former code.
686:
687: The output is very simple: only an estimate of the intercept and of
688: the slope with 95% confident intervals.
689:
690: Current limitations:
691: A) Even if you enter covariates, i.e. with the
692: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
693: B) There is no computation of Life Expectancy nor Life Table.
694:
695: Revision 1.97 2004/02/20 13:25:42 lievre
696: Version 0.96d. Population forecasting command line is (temporarily)
697: suppressed.
698:
699: Revision 1.96 2003/07/15 15:38:55 brouard
700: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
701: rewritten within the same printf. Workaround: many printfs.
702:
703: Revision 1.95 2003/07/08 07:54:34 brouard
704: * imach.c (Repository):
705: (Repository): Using imachwizard code to output a more meaningful covariance
706: matrix (cov(a12,c31) instead of numbers.
707:
708: Revision 1.94 2003/06/27 13:00:02 brouard
709: Just cleaning
710:
711: Revision 1.93 2003/06/25 16:33:55 brouard
712: (Module): On windows (cygwin) function asctime_r doesn't
713: exist so I changed back to asctime which exists.
714: (Module): Version 0.96b
715:
716: Revision 1.92 2003/06/25 16:30:45 brouard
717: (Module): On windows (cygwin) function asctime_r doesn't
718: exist so I changed back to asctime which exists.
719:
720: Revision 1.91 2003/06/25 15:30:29 brouard
721: * imach.c (Repository): Duplicated warning errors corrected.
722: (Repository): Elapsed time after each iteration is now output. It
723: helps to forecast when convergence will be reached. Elapsed time
724: is stamped in powell. We created a new html file for the graphs
725: concerning matrix of covariance. It has extension -cov.htm.
726:
727: Revision 1.90 2003/06/24 12:34:15 brouard
728: (Module): Some bugs corrected for windows. Also, when
729: mle=-1 a template is output in file "or"mypar.txt with the design
730: of the covariance matrix to be input.
731:
732: Revision 1.89 2003/06/24 12:30:52 brouard
733: (Module): Some bugs corrected for windows. Also, when
734: mle=-1 a template is output in file "or"mypar.txt with the design
735: of the covariance matrix to be input.
736:
737: Revision 1.88 2003/06/23 17:54:56 brouard
738: * 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.
739:
740: Revision 1.87 2003/06/18 12:26:01 brouard
741: Version 0.96
742:
743: Revision 1.86 2003/06/17 20:04:08 brouard
744: (Module): Change position of html and gnuplot routines and added
745: routine fileappend.
746:
747: Revision 1.85 2003/06/17 13:12:43 brouard
748: * imach.c (Repository): Check when date of death was earlier that
749: current date of interview. It may happen when the death was just
750: prior to the death. In this case, dh was negative and likelihood
751: was wrong (infinity). We still send an "Error" but patch by
752: assuming that the date of death was just one stepm after the
753: interview.
754: (Repository): Because some people have very long ID (first column)
755: we changed int to long in num[] and we added a new lvector for
756: memory allocation. But we also truncated to 8 characters (left
757: truncation)
758: (Repository): No more line truncation errors.
759:
760: Revision 1.84 2003/06/13 21:44:43 brouard
761: * imach.c (Repository): Replace "freqsummary" at a correct
762: place. It differs from routine "prevalence" which may be called
763: many times. Probs is memory consuming and must be used with
764: parcimony.
765: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
766:
767: Revision 1.83 2003/06/10 13:39:11 lievre
768: *** empty log message ***
769:
770: Revision 1.82 2003/06/05 15:57:20 brouard
771: Add log in imach.c and fullversion number is now printed.
772:
773: */
774: /*
775: Interpolated Markov Chain
776:
777: Short summary of the programme:
778:
1.227 brouard 779: This program computes Healthy Life Expectancies or State-specific
780: (if states aren't health statuses) Expectancies from
781: cross-longitudinal data. Cross-longitudinal data consist in:
782:
783: -1- a first survey ("cross") where individuals from different ages
784: are interviewed on their health status or degree of disability (in
785: the case of a health survey which is our main interest)
786:
787: -2- at least a second wave of interviews ("longitudinal") which
788: measure each change (if any) in individual health status. Health
789: expectancies are computed from the time spent in each health state
790: according to a model. More health states you consider, more time is
791: necessary to reach the Maximum Likelihood of the parameters involved
792: in the model. The simplest model is the multinomial logistic model
793: where pij is the probability to be observed in state j at the second
794: wave conditional to be observed in state i at the first
795: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
796: etc , where 'age' is age and 'sex' is a covariate. If you want to
797: have a more complex model than "constant and age", you should modify
798: the program where the markup *Covariates have to be included here
799: again* invites you to do it. More covariates you add, slower the
1.126 brouard 800: convergence.
801:
802: The advantage of this computer programme, compared to a simple
803: multinomial logistic model, is clear when the delay between waves is not
804: identical for each individual. Also, if a individual missed an
805: intermediate interview, the information is lost, but taken into
806: account using an interpolation or extrapolation.
807:
808: hPijx is the probability to be observed in state i at age x+h
809: conditional to the observed state i at age x. The delay 'h' can be
810: split into an exact number (nh*stepm) of unobserved intermediate
811: states. This elementary transition (by month, quarter,
812: semester or year) is modelled as a multinomial logistic. The hPx
813: matrix is simply the matrix product of nh*stepm elementary matrices
814: and the contribution of each individual to the likelihood is simply
815: hPijx.
816:
817: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 818: of the life expectancies. It also computes the period (stable) prevalence.
819:
820: Back prevalence and projections:
1.227 brouard 821:
822: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
823: double agemaxpar, double ftolpl, int *ncvyearp, double
824: dateprev1,double dateprev2, int firstpass, int lastpass, int
825: mobilavproj)
826:
827: Computes the back prevalence limit for any combination of
828: covariate values k at any age between ageminpar and agemaxpar and
829: returns it in **bprlim. In the loops,
830:
831: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
832: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
833:
834: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 835: Computes for any combination of covariates k and any age between bage and fage
836: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
837: oldm=oldms;savm=savms;
1.227 brouard 838:
1.267 brouard 839: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 840: Computes the transition matrix starting at age 'age' over
841: 'nhstepm*hstepm*stepm' months (i.e. until
842: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 843: nhstepm*hstepm matrices.
844:
845: Returns p3mat[i][j][h] after calling
846: p3mat[i][j][h]=matprod2(newm,
847: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
848: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
849: oldm);
1.226 brouard 850:
851: Important routines
852:
853: - func (or funcone), computes logit (pij) distinguishing
854: o fixed variables (single or product dummies or quantitative);
855: o varying variables by:
856: (1) wave (single, product dummies, quantitative),
857: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
858: % fixed dummy (treated) or quantitative (not done because time-consuming);
859: % varying dummy (not done) or quantitative (not done);
860: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
861: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
862: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
863: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
864: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 865:
1.226 brouard 866:
867:
1.133 brouard 868: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
869: Institut national d'études démographiques, Paris.
1.126 brouard 870: This software have been partly granted by Euro-REVES, a concerted action
871: from the European Union.
872: It is copyrighted identically to a GNU software product, ie programme and
873: software can be distributed freely for non commercial use. Latest version
874: can be accessed at http://euroreves.ined.fr/imach .
875:
876: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
877: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
878:
879: **********************************************************************/
880: /*
881: main
882: read parameterfile
883: read datafile
884: concatwav
885: freqsummary
886: if (mle >= 1)
887: mlikeli
888: print results files
889: if mle==1
890: computes hessian
891: read end of parameter file: agemin, agemax, bage, fage, estepm
892: begin-prev-date,...
893: open gnuplot file
894: open html file
1.145 brouard 895: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
896: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
897: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
898: freexexit2 possible for memory heap.
899:
900: h Pij x | pij_nom ficrestpij
901: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
902: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
903: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
904:
905: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
906: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
907: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
908: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
909: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
910:
1.126 brouard 911: forecasting if prevfcast==1 prevforecast call prevalence()
912: health expectancies
913: Variance-covariance of DFLE
914: prevalence()
915: movingaverage()
916: varevsij()
917: if popbased==1 varevsij(,popbased)
918: total life expectancies
919: Variance of period (stable) prevalence
920: end
921: */
922:
1.187 brouard 923: /* #define DEBUG */
924: /* #define DEBUGBRENT */
1.203 brouard 925: /* #define DEBUGLINMIN */
926: /* #define DEBUGHESS */
927: #define DEBUGHESSIJ
1.224 brouard 928: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 929: #define POWELL /* Instead of NLOPT */
1.224 brouard 930: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 931: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
932: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 933:
934: #include <math.h>
935: #include <stdio.h>
936: #include <stdlib.h>
937: #include <string.h>
1.226 brouard 938: #include <ctype.h>
1.159 brouard 939:
940: #ifdef _WIN32
941: #include <io.h>
1.172 brouard 942: #include <windows.h>
943: #include <tchar.h>
1.159 brouard 944: #else
1.126 brouard 945: #include <unistd.h>
1.159 brouard 946: #endif
1.126 brouard 947:
948: #include <limits.h>
949: #include <sys/types.h>
1.171 brouard 950:
951: #if defined(__GNUC__)
952: #include <sys/utsname.h> /* Doesn't work on Windows */
953: #endif
954:
1.126 brouard 955: #include <sys/stat.h>
956: #include <errno.h>
1.159 brouard 957: /* extern int errno; */
1.126 brouard 958:
1.157 brouard 959: /* #ifdef LINUX */
960: /* #include <time.h> */
961: /* #include "timeval.h" */
962: /* #else */
963: /* #include <sys/time.h> */
964: /* #endif */
965:
1.126 brouard 966: #include <time.h>
967:
1.136 brouard 968: #ifdef GSL
969: #include <gsl/gsl_errno.h>
970: #include <gsl/gsl_multimin.h>
971: #endif
972:
1.167 brouard 973:
1.162 brouard 974: #ifdef NLOPT
975: #include <nlopt.h>
976: typedef struct {
977: double (* function)(double [] );
978: } myfunc_data ;
979: #endif
980:
1.126 brouard 981: /* #include <libintl.h> */
982: /* #define _(String) gettext (String) */
983:
1.251 brouard 984: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 985:
986: #define GNUPLOTPROGRAM "gnuplot"
987: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
988: #define FILENAMELENGTH 132
989:
990: #define GLOCK_ERROR_NOPATH -1 /* empty path */
991: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
992:
1.144 brouard 993: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
994: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 995:
996: #define NINTERVMAX 8
1.144 brouard 997: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
998: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
999: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1000: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1001: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1002: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1003: #define MAXN 20000
1.144 brouard 1004: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1005: /* #define AGESUP 130 */
1006: #define AGESUP 150
1.268 brouard 1007: #define AGEINF 0
1.218 brouard 1008: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1009: #define AGEBASE 40
1.194 brouard 1010: #define AGEOVERFLOW 1.e20
1.164 brouard 1011: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1012: #ifdef _WIN32
1013: #define DIRSEPARATOR '\\'
1014: #define CHARSEPARATOR "\\"
1015: #define ODIRSEPARATOR '/'
1016: #else
1.126 brouard 1017: #define DIRSEPARATOR '/'
1018: #define CHARSEPARATOR "/"
1019: #define ODIRSEPARATOR '\\'
1020: #endif
1021:
1.273 ! brouard 1022: /* $Id: imach.c,v 1.272 2017/06/27 10:22:40 brouard Exp $ */
1.126 brouard 1023: /* $State: Exp $ */
1.196 brouard 1024: #include "version.h"
1025: char version[]=__IMACH_VERSION__;
1.224 brouard 1026: 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";
1.273 ! brouard 1027: char fullversion[]="$Revision: 1.272 $ $Date: 2017/06/27 10:22:40 $";
1.126 brouard 1028: char strstart[80];
1029: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1030: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1031: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1032: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1033: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1034: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1035: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1036: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1037: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1038: int cptcovprodnoage=0; /**< Number of covariate products without age */
1039: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1040: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1041: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1042: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1043: int nsd=0; /**< Total number of single dummy variables (output) */
1044: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1045: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1046: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1047: int ntveff=0; /**< ntveff number of effective time varying variables */
1048: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1049: int cptcov=0; /* Working variable */
1.218 brouard 1050: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1051: int npar=NPARMAX;
1052: int nlstate=2; /* Number of live states */
1053: int ndeath=1; /* Number of dead states */
1.130 brouard 1054: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1055: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1056: int popbased=0;
1057:
1058: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1059: int maxwav=0; /* Maxim number of waves */
1060: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1061: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1062: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1063: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1064: int mle=1, weightopt=0;
1.126 brouard 1065: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1066: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1067: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1068: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1069: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1070: int selected(int kvar); /* Is covariate kvar selected for printing results */
1071:
1.130 brouard 1072: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1073: double **matprod2(); /* test */
1.126 brouard 1074: double **oldm, **newm, **savm; /* Working pointers to matrices */
1075: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1076: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1077:
1.136 brouard 1078: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1079: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1080: FILE *ficlog, *ficrespow;
1.130 brouard 1081: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1082: double fretone; /* Only one call to likelihood */
1.130 brouard 1083: long ipmx=0; /* Number of contributions */
1.126 brouard 1084: double sw; /* Sum of weights */
1085: char filerespow[FILENAMELENGTH];
1086: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1087: FILE *ficresilk;
1088: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1089: FILE *ficresprobmorprev;
1090: FILE *fichtm, *fichtmcov; /* Html File */
1091: FILE *ficreseij;
1092: char filerese[FILENAMELENGTH];
1093: FILE *ficresstdeij;
1094: char fileresstde[FILENAMELENGTH];
1095: FILE *ficrescveij;
1096: char filerescve[FILENAMELENGTH];
1097: FILE *ficresvij;
1098: char fileresv[FILENAMELENGTH];
1.269 brouard 1099:
1.126 brouard 1100: char title[MAXLINE];
1.234 brouard 1101: char model[MAXLINE]; /**< The model line */
1.217 brouard 1102: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1103: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1104: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1105: char command[FILENAMELENGTH];
1106: int outcmd=0;
1107:
1.217 brouard 1108: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1109: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1110: char filelog[FILENAMELENGTH]; /* Log file */
1111: char filerest[FILENAMELENGTH];
1112: char fileregp[FILENAMELENGTH];
1113: char popfile[FILENAMELENGTH];
1114:
1115: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1116:
1.157 brouard 1117: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1118: /* struct timezone tzp; */
1119: /* extern int gettimeofday(); */
1120: struct tm tml, *gmtime(), *localtime();
1121:
1122: extern time_t time();
1123:
1124: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1125: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1126: struct tm tm;
1127:
1.126 brouard 1128: char strcurr[80], strfor[80];
1129:
1130: char *endptr;
1131: long lval;
1132: double dval;
1133:
1134: #define NR_END 1
1135: #define FREE_ARG char*
1136: #define FTOL 1.0e-10
1137:
1138: #define NRANSI
1.240 brouard 1139: #define ITMAX 200
1140: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1141:
1142: #define TOL 2.0e-4
1143:
1144: #define CGOLD 0.3819660
1145: #define ZEPS 1.0e-10
1146: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1147:
1148: #define GOLD 1.618034
1149: #define GLIMIT 100.0
1150: #define TINY 1.0e-20
1151:
1152: static double maxarg1,maxarg2;
1153: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1154: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1155:
1156: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1157: #define rint(a) floor(a+0.5)
1.166 brouard 1158: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1159: #define mytinydouble 1.0e-16
1.166 brouard 1160: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1161: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1162: /* static double dsqrarg; */
1163: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1164: static double sqrarg;
1165: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1166: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1167: int agegomp= AGEGOMP;
1168:
1169: int imx;
1170: int stepm=1;
1171: /* Stepm, step in month: minimum step interpolation*/
1172:
1173: int estepm;
1174: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1175:
1176: int m,nb;
1177: long *num;
1.197 brouard 1178: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1179: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1180: covariate for which somebody answered excluding
1181: undefined. Usually 2: 0 and 1. */
1182: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1183: covariate for which somebody answered including
1184: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1185: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1186: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1187: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1188: double *ageexmed,*agecens;
1189: double dateintmean=0;
1190:
1191: double *weight;
1192: int **s; /* Status */
1.141 brouard 1193: double *agedc;
1.145 brouard 1194: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1195: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1196: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1197: double **coqvar; /* Fixed quantitative covariate nqv */
1198: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1199: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1200: double idx;
1201: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1202: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1203: /*k 1 2 3 4 5 6 7 8 9 */
1204: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1205: /* Tndvar[k] 1 2 3 4 5 */
1206: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1207: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1208: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1209: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1210: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1211: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1212: /* Tprod[i]=k 4 7 */
1213: /* Tage[i]=k 5 8 */
1214: /* */
1215: /* Type */
1216: /* V 1 2 3 4 5 */
1217: /* F F V V V */
1218: /* D Q D D Q */
1219: /* */
1220: int *TvarsD;
1221: int *TvarsDind;
1222: int *TvarsQ;
1223: int *TvarsQind;
1224:
1.235 brouard 1225: #define MAXRESULTLINES 10
1226: int nresult=0;
1.258 brouard 1227: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1228: int TKresult[MAXRESULTLINES];
1.237 brouard 1229: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1230: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1231: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1232: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1233: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1234: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1235:
1.234 brouard 1236: /* 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 1237: 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 */
1238: 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 */
1239: 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 */
1240: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1241: 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 */
1242: 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 1243: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1244: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1245: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1246: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1247: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1248: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1249: 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 */
1250: 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 */
1251:
1.230 brouard 1252: int *Tvarsel; /**< Selected covariates for output */
1253: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1254: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1255: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1256: 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 1257: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1258: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1259: int *Tage;
1.227 brouard 1260: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1261: 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 1262: 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*/
1263: 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 1264: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1265: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1266: int **Tvard;
1267: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1268: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1269: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1270: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1271: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1272: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1273: double *lsurv, *lpop, *tpop;
1274:
1.231 brouard 1275: #define FD 1; /* Fixed dummy covariate */
1276: #define FQ 2; /* Fixed quantitative covariate */
1277: #define FP 3; /* Fixed product covariate */
1278: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1279: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1280: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1281: #define VD 10; /* Varying dummy covariate */
1282: #define VQ 11; /* Varying quantitative covariate */
1283: #define VP 12; /* Varying product covariate */
1284: #define VPDD 13; /* Varying product dummy*dummy covariate */
1285: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1286: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1287: #define APFD 16; /* Age product * fixed dummy covariate */
1288: #define APFQ 17; /* Age product * fixed quantitative covariate */
1289: #define APVD 18; /* Age product * varying dummy covariate */
1290: #define APVQ 19; /* Age product * varying quantitative covariate */
1291:
1292: #define FTYPE 1; /* Fixed covariate */
1293: #define VTYPE 2; /* Varying covariate (loop in wave) */
1294: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1295:
1296: struct kmodel{
1297: int maintype; /* main type */
1298: int subtype; /* subtype */
1299: };
1300: struct kmodel modell[NCOVMAX];
1301:
1.143 brouard 1302: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1303: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1304:
1305: /**************** split *************************/
1306: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1307: {
1308: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1309: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1310: */
1311: char *ss; /* pointer */
1.186 brouard 1312: int l1=0, l2=0; /* length counters */
1.126 brouard 1313:
1314: l1 = strlen(path ); /* length of path */
1315: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1316: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1317: if ( ss == NULL ) { /* no directory, so determine current directory */
1318: strcpy( name, path ); /* we got the fullname name because no directory */
1319: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1320: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1321: /* get current working directory */
1322: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1323: #ifdef WIN32
1324: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1325: #else
1326: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1327: #endif
1.126 brouard 1328: return( GLOCK_ERROR_GETCWD );
1329: }
1330: /* got dirc from getcwd*/
1331: printf(" DIRC = %s \n",dirc);
1.205 brouard 1332: } else { /* strip directory from path */
1.126 brouard 1333: ss++; /* after this, the filename */
1334: l2 = strlen( ss ); /* length of filename */
1335: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1336: strcpy( name, ss ); /* save file name */
1337: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1338: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1339: printf(" DIRC2 = %s \n",dirc);
1340: }
1341: /* We add a separator at the end of dirc if not exists */
1342: l1 = strlen( dirc ); /* length of directory */
1343: if( dirc[l1-1] != DIRSEPARATOR ){
1344: dirc[l1] = DIRSEPARATOR;
1345: dirc[l1+1] = 0;
1346: printf(" DIRC3 = %s \n",dirc);
1347: }
1348: ss = strrchr( name, '.' ); /* find last / */
1349: if (ss >0){
1350: ss++;
1351: strcpy(ext,ss); /* save extension */
1352: l1= strlen( name);
1353: l2= strlen(ss)+1;
1354: strncpy( finame, name, l1-l2);
1355: finame[l1-l2]= 0;
1356: }
1357:
1358: return( 0 ); /* we're done */
1359: }
1360:
1361:
1362: /******************************************/
1363:
1364: void replace_back_to_slash(char *s, char*t)
1365: {
1366: int i;
1367: int lg=0;
1368: i=0;
1369: lg=strlen(t);
1370: for(i=0; i<= lg; i++) {
1371: (s[i] = t[i]);
1372: if (t[i]== '\\') s[i]='/';
1373: }
1374: }
1375:
1.132 brouard 1376: char *trimbb(char *out, char *in)
1.137 brouard 1377: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1378: char *s;
1379: s=out;
1380: while (*in != '\0'){
1.137 brouard 1381: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1382: in++;
1383: }
1384: *out++ = *in++;
1385: }
1386: *out='\0';
1387: return s;
1388: }
1389:
1.187 brouard 1390: /* char *substrchaine(char *out, char *in, char *chain) */
1391: /* { */
1392: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1393: /* char *s, *t; */
1394: /* t=in;s=out; */
1395: /* while ((*in != *chain) && (*in != '\0')){ */
1396: /* *out++ = *in++; */
1397: /* } */
1398:
1399: /* /\* *in matches *chain *\/ */
1400: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1401: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1402: /* } */
1403: /* in--; chain--; */
1404: /* while ( (*in != '\0')){ */
1405: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1406: /* *out++ = *in++; */
1407: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1408: /* } */
1409: /* *out='\0'; */
1410: /* out=s; */
1411: /* return out; */
1412: /* } */
1413: char *substrchaine(char *out, char *in, char *chain)
1414: {
1415: /* Substract chain 'chain' from 'in', return and output 'out' */
1416: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1417:
1418: char *strloc;
1419:
1420: strcpy (out, in);
1421: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1422: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1423: if(strloc != NULL){
1424: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1425: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1426: /* strcpy (strloc, strloc +strlen(chain));*/
1427: }
1428: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1429: return out;
1430: }
1431:
1432:
1.145 brouard 1433: char *cutl(char *blocc, char *alocc, char *in, char occ)
1434: {
1.187 brouard 1435: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1436: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1437: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1438: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1439: */
1.160 brouard 1440: char *s, *t;
1.145 brouard 1441: t=in;s=in;
1442: while ((*in != occ) && (*in != '\0')){
1443: *alocc++ = *in++;
1444: }
1445: if( *in == occ){
1446: *(alocc)='\0';
1447: s=++in;
1448: }
1449:
1450: if (s == t) {/* occ not found */
1451: *(alocc-(in-s))='\0';
1452: in=s;
1453: }
1454: while ( *in != '\0'){
1455: *blocc++ = *in++;
1456: }
1457:
1458: *blocc='\0';
1459: return t;
1460: }
1.137 brouard 1461: char *cutv(char *blocc, char *alocc, char *in, char occ)
1462: {
1.187 brouard 1463: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1464: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1465: gives blocc="abcdef2ghi" and alocc="j".
1466: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1467: */
1468: char *s, *t;
1469: t=in;s=in;
1470: while (*in != '\0'){
1471: while( *in == occ){
1472: *blocc++ = *in++;
1473: s=in;
1474: }
1475: *blocc++ = *in++;
1476: }
1477: if (s == t) /* occ not found */
1478: *(blocc-(in-s))='\0';
1479: else
1480: *(blocc-(in-s)-1)='\0';
1481: in=s;
1482: while ( *in != '\0'){
1483: *alocc++ = *in++;
1484: }
1485:
1486: *alocc='\0';
1487: return s;
1488: }
1489:
1.126 brouard 1490: int nbocc(char *s, char occ)
1491: {
1492: int i,j=0;
1493: int lg=20;
1494: i=0;
1495: lg=strlen(s);
1496: for(i=0; i<= lg; i++) {
1.234 brouard 1497: if (s[i] == occ ) j++;
1.126 brouard 1498: }
1499: return j;
1500: }
1501:
1.137 brouard 1502: /* void cutv(char *u,char *v, char*t, char occ) */
1503: /* { */
1504: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1505: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1506: /* gives u="abcdef2ghi" and v="j" *\/ */
1507: /* int i,lg,j,p=0; */
1508: /* i=0; */
1509: /* lg=strlen(t); */
1510: /* for(j=0; j<=lg-1; j++) { */
1511: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1512: /* } */
1.126 brouard 1513:
1.137 brouard 1514: /* for(j=0; j<p; j++) { */
1515: /* (u[j] = t[j]); */
1516: /* } */
1517: /* u[p]='\0'; */
1.126 brouard 1518:
1.137 brouard 1519: /* for(j=0; j<= lg; j++) { */
1520: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1521: /* } */
1522: /* } */
1.126 brouard 1523:
1.160 brouard 1524: #ifdef _WIN32
1525: char * strsep(char **pp, const char *delim)
1526: {
1527: char *p, *q;
1528:
1529: if ((p = *pp) == NULL)
1530: return 0;
1531: if ((q = strpbrk (p, delim)) != NULL)
1532: {
1533: *pp = q + 1;
1534: *q = '\0';
1535: }
1536: else
1537: *pp = 0;
1538: return p;
1539: }
1540: #endif
1541:
1.126 brouard 1542: /********************** nrerror ********************/
1543:
1544: void nrerror(char error_text[])
1545: {
1546: fprintf(stderr,"ERREUR ...\n");
1547: fprintf(stderr,"%s\n",error_text);
1548: exit(EXIT_FAILURE);
1549: }
1550: /*********************** vector *******************/
1551: double *vector(int nl, int nh)
1552: {
1553: double *v;
1554: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1555: if (!v) nrerror("allocation failure in vector");
1556: return v-nl+NR_END;
1557: }
1558:
1559: /************************ free vector ******************/
1560: void free_vector(double*v, int nl, int nh)
1561: {
1562: free((FREE_ARG)(v+nl-NR_END));
1563: }
1564:
1565: /************************ivector *******************************/
1566: int *ivector(long nl,long nh)
1567: {
1568: int *v;
1569: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1570: if (!v) nrerror("allocation failure in ivector");
1571: return v-nl+NR_END;
1572: }
1573:
1574: /******************free ivector **************************/
1575: void free_ivector(int *v, long nl, long nh)
1576: {
1577: free((FREE_ARG)(v+nl-NR_END));
1578: }
1579:
1580: /************************lvector *******************************/
1581: long *lvector(long nl,long nh)
1582: {
1583: long *v;
1584: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1585: if (!v) nrerror("allocation failure in ivector");
1586: return v-nl+NR_END;
1587: }
1588:
1589: /******************free lvector **************************/
1590: void free_lvector(long *v, long nl, long nh)
1591: {
1592: free((FREE_ARG)(v+nl-NR_END));
1593: }
1594:
1595: /******************* imatrix *******************************/
1596: int **imatrix(long nrl, long nrh, long ncl, long nch)
1597: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1598: {
1599: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1600: int **m;
1601:
1602: /* allocate pointers to rows */
1603: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1604: if (!m) nrerror("allocation failure 1 in matrix()");
1605: m += NR_END;
1606: m -= nrl;
1607:
1608:
1609: /* allocate rows and set pointers to them */
1610: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1611: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1612: m[nrl] += NR_END;
1613: m[nrl] -= ncl;
1614:
1615: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1616:
1617: /* return pointer to array of pointers to rows */
1618: return m;
1619: }
1620:
1621: /****************** free_imatrix *************************/
1622: void free_imatrix(m,nrl,nrh,ncl,nch)
1623: int **m;
1624: long nch,ncl,nrh,nrl;
1625: /* free an int matrix allocated by imatrix() */
1626: {
1627: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1628: free((FREE_ARG) (m+nrl-NR_END));
1629: }
1630:
1631: /******************* matrix *******************************/
1632: double **matrix(long nrl, long nrh, long ncl, long nch)
1633: {
1634: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1635: double **m;
1636:
1637: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1638: if (!m) nrerror("allocation failure 1 in matrix()");
1639: m += NR_END;
1640: m -= nrl;
1641:
1642: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1643: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1644: m[nrl] += NR_END;
1645: m[nrl] -= ncl;
1646:
1647: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1648: return m;
1.145 brouard 1649: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1650: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1651: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1652: */
1653: }
1654:
1655: /*************************free matrix ************************/
1656: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1657: {
1658: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1659: free((FREE_ARG)(m+nrl-NR_END));
1660: }
1661:
1662: /******************* ma3x *******************************/
1663: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1664: {
1665: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1666: double ***m;
1667:
1668: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1669: if (!m) nrerror("allocation failure 1 in matrix()");
1670: m += NR_END;
1671: m -= nrl;
1672:
1673: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1674: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1675: m[nrl] += NR_END;
1676: m[nrl] -= ncl;
1677:
1678: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1679:
1680: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1681: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1682: m[nrl][ncl] += NR_END;
1683: m[nrl][ncl] -= nll;
1684: for (j=ncl+1; j<=nch; j++)
1685: m[nrl][j]=m[nrl][j-1]+nlay;
1686:
1687: for (i=nrl+1; i<=nrh; i++) {
1688: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1689: for (j=ncl+1; j<=nch; j++)
1690: m[i][j]=m[i][j-1]+nlay;
1691: }
1692: return m;
1693: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1694: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1695: */
1696: }
1697:
1698: /*************************free ma3x ************************/
1699: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1700: {
1701: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1702: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1703: free((FREE_ARG)(m+nrl-NR_END));
1704: }
1705:
1706: /*************** function subdirf ***********/
1707: char *subdirf(char fileres[])
1708: {
1709: /* Caution optionfilefiname is hidden */
1710: strcpy(tmpout,optionfilefiname);
1711: strcat(tmpout,"/"); /* Add to the right */
1712: strcat(tmpout,fileres);
1713: return tmpout;
1714: }
1715:
1716: /*************** function subdirf2 ***********/
1717: char *subdirf2(char fileres[], char *preop)
1718: {
1719:
1720: /* Caution optionfilefiname is hidden */
1721: strcpy(tmpout,optionfilefiname);
1722: strcat(tmpout,"/");
1723: strcat(tmpout,preop);
1724: strcat(tmpout,fileres);
1725: return tmpout;
1726: }
1727:
1728: /*************** function subdirf3 ***********/
1729: char *subdirf3(char fileres[], char *preop, char *preop2)
1730: {
1731:
1732: /* Caution optionfilefiname is hidden */
1733: strcpy(tmpout,optionfilefiname);
1734: strcat(tmpout,"/");
1735: strcat(tmpout,preop);
1736: strcat(tmpout,preop2);
1737: strcat(tmpout,fileres);
1738: return tmpout;
1739: }
1.213 brouard 1740:
1741: /*************** function subdirfext ***********/
1742: char *subdirfext(char fileres[], char *preop, char *postop)
1743: {
1744:
1745: strcpy(tmpout,preop);
1746: strcat(tmpout,fileres);
1747: strcat(tmpout,postop);
1748: return tmpout;
1749: }
1.126 brouard 1750:
1.213 brouard 1751: /*************** function subdirfext3 ***********/
1752: char *subdirfext3(char fileres[], char *preop, char *postop)
1753: {
1754:
1755: /* Caution optionfilefiname is hidden */
1756: strcpy(tmpout,optionfilefiname);
1757: strcat(tmpout,"/");
1758: strcat(tmpout,preop);
1759: strcat(tmpout,fileres);
1760: strcat(tmpout,postop);
1761: return tmpout;
1762: }
1763:
1.162 brouard 1764: char *asc_diff_time(long time_sec, char ascdiff[])
1765: {
1766: long sec_left, days, hours, minutes;
1767: days = (time_sec) / (60*60*24);
1768: sec_left = (time_sec) % (60*60*24);
1769: hours = (sec_left) / (60*60) ;
1770: sec_left = (sec_left) %(60*60);
1771: minutes = (sec_left) /60;
1772: sec_left = (sec_left) % (60);
1773: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1774: return ascdiff;
1775: }
1776:
1.126 brouard 1777: /***************** f1dim *************************/
1778: extern int ncom;
1779: extern double *pcom,*xicom;
1780: extern double (*nrfunc)(double []);
1781:
1782: double f1dim(double x)
1783: {
1784: int j;
1785: double f;
1786: double *xt;
1787:
1788: xt=vector(1,ncom);
1789: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1790: f=(*nrfunc)(xt);
1791: free_vector(xt,1,ncom);
1792: return f;
1793: }
1794:
1795: /*****************brent *************************/
1796: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1797: {
1798: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1799: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1800: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1801: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1802: * returned function value.
1803: */
1.126 brouard 1804: int iter;
1805: double a,b,d,etemp;
1.159 brouard 1806: double fu=0,fv,fw,fx;
1.164 brouard 1807: double ftemp=0.;
1.126 brouard 1808: double p,q,r,tol1,tol2,u,v,w,x,xm;
1809: double e=0.0;
1810:
1811: a=(ax < cx ? ax : cx);
1812: b=(ax > cx ? ax : cx);
1813: x=w=v=bx;
1814: fw=fv=fx=(*f)(x);
1815: for (iter=1;iter<=ITMAX;iter++) {
1816: xm=0.5*(a+b);
1817: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1818: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1819: printf(".");fflush(stdout);
1820: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1821: #ifdef DEBUGBRENT
1.126 brouard 1822: 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);
1823: 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);
1824: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1825: #endif
1826: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1827: *xmin=x;
1828: return fx;
1829: }
1830: ftemp=fu;
1831: if (fabs(e) > tol1) {
1832: r=(x-w)*(fx-fv);
1833: q=(x-v)*(fx-fw);
1834: p=(x-v)*q-(x-w)*r;
1835: q=2.0*(q-r);
1836: if (q > 0.0) p = -p;
1837: q=fabs(q);
1838: etemp=e;
1839: e=d;
1840: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1841: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1842: else {
1.224 brouard 1843: d=p/q;
1844: u=x+d;
1845: if (u-a < tol2 || b-u < tol2)
1846: d=SIGN(tol1,xm-x);
1.126 brouard 1847: }
1848: } else {
1849: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1850: }
1851: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1852: fu=(*f)(u);
1853: if (fu <= fx) {
1854: if (u >= x) a=x; else b=x;
1855: SHFT(v,w,x,u)
1.183 brouard 1856: SHFT(fv,fw,fx,fu)
1857: } else {
1858: if (u < x) a=u; else b=u;
1859: if (fu <= fw || w == x) {
1.224 brouard 1860: v=w;
1861: w=u;
1862: fv=fw;
1863: fw=fu;
1.183 brouard 1864: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1865: v=u;
1866: fv=fu;
1.183 brouard 1867: }
1868: }
1.126 brouard 1869: }
1870: nrerror("Too many iterations in brent");
1871: *xmin=x;
1872: return fx;
1873: }
1874:
1875: /****************** mnbrak ***********************/
1876:
1877: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1878: double (*func)(double))
1.183 brouard 1879: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1880: the downhill direction (defined by the function as evaluated at the initial points) and returns
1881: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1882: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1883: */
1.126 brouard 1884: double ulim,u,r,q, dum;
1885: double fu;
1.187 brouard 1886:
1887: double scale=10.;
1888: int iterscale=0;
1889:
1890: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1891: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1892:
1893:
1894: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1895: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1896: /* *bx = *ax - (*ax - *bx)/scale; */
1897: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1898: /* } */
1899:
1.126 brouard 1900: if (*fb > *fa) {
1901: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1902: SHFT(dum,*fb,*fa,dum)
1903: }
1.126 brouard 1904: *cx=(*bx)+GOLD*(*bx-*ax);
1905: *fc=(*func)(*cx);
1.183 brouard 1906: #ifdef DEBUG
1.224 brouard 1907: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1908: 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 1909: #endif
1.224 brouard 1910: 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 1911: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1912: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1913: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1914: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1915: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1916: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1917: fu=(*func)(u);
1.163 brouard 1918: #ifdef DEBUG
1919: /* f(x)=A(x-u)**2+f(u) */
1920: double A, fparabu;
1921: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1922: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1923: 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);
1924: 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 1925: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1926: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1927: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1928: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1929: #endif
1.184 brouard 1930: #ifdef MNBRAKORIGINAL
1.183 brouard 1931: #else
1.191 brouard 1932: /* if (fu > *fc) { */
1933: /* #ifdef DEBUG */
1934: /* printf("mnbrak4 fu > fc \n"); */
1935: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1936: /* #endif */
1937: /* /\* 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 *\\/ *\/ */
1938: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1939: /* dum=u; /\* Shifting c and u *\/ */
1940: /* u = *cx; */
1941: /* *cx = dum; */
1942: /* dum = fu; */
1943: /* fu = *fc; */
1944: /* *fc =dum; */
1945: /* } else { /\* end *\/ */
1946: /* #ifdef DEBUG */
1947: /* printf("mnbrak3 fu < fc \n"); */
1948: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1949: /* #endif */
1950: /* dum=u; /\* Shifting c and u *\/ */
1951: /* u = *cx; */
1952: /* *cx = dum; */
1953: /* dum = fu; */
1954: /* fu = *fc; */
1955: /* *fc =dum; */
1956: /* } */
1.224 brouard 1957: #ifdef DEBUGMNBRAK
1958: double A, fparabu;
1959: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1960: fparabu= *fa - A*(*ax-u)*(*ax-u);
1961: 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);
1962: 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 1963: #endif
1.191 brouard 1964: dum=u; /* Shifting c and u */
1965: u = *cx;
1966: *cx = dum;
1967: dum = fu;
1968: fu = *fc;
1969: *fc =dum;
1.183 brouard 1970: #endif
1.162 brouard 1971: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1972: #ifdef DEBUG
1.224 brouard 1973: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1974: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1975: #endif
1.126 brouard 1976: fu=(*func)(u);
1977: if (fu < *fc) {
1.183 brouard 1978: #ifdef DEBUG
1.224 brouard 1979: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1980: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1981: #endif
1982: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1983: SHFT(*fb,*fc,fu,(*func)(u))
1984: #ifdef DEBUG
1985: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1986: #endif
1987: }
1.162 brouard 1988: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1989: #ifdef DEBUG
1.224 brouard 1990: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1991: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1992: #endif
1.126 brouard 1993: u=ulim;
1994: fu=(*func)(u);
1.183 brouard 1995: } else { /* u could be left to b (if r > q parabola has a maximum) */
1996: #ifdef DEBUG
1.224 brouard 1997: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1998: 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 1999: #endif
1.126 brouard 2000: u=(*cx)+GOLD*(*cx-*bx);
2001: fu=(*func)(u);
1.224 brouard 2002: #ifdef DEBUG
2003: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2004: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2005: #endif
1.183 brouard 2006: } /* end tests */
1.126 brouard 2007: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2008: SHFT(*fa,*fb,*fc,fu)
2009: #ifdef DEBUG
1.224 brouard 2010: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2011: 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 2012: #endif
2013: } /* 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 2014: }
2015:
2016: /*************** linmin ************************/
1.162 brouard 2017: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2018: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2019: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2020: the value of func at the returned location p . This is actually all accomplished by calling the
2021: routines mnbrak and brent .*/
1.126 brouard 2022: int ncom;
2023: double *pcom,*xicom;
2024: double (*nrfunc)(double []);
2025:
1.224 brouard 2026: #ifdef LINMINORIGINAL
1.126 brouard 2027: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2028: #else
2029: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2030: #endif
1.126 brouard 2031: {
2032: double brent(double ax, double bx, double cx,
2033: double (*f)(double), double tol, double *xmin);
2034: double f1dim(double x);
2035: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2036: double *fc, double (*func)(double));
2037: int j;
2038: double xx,xmin,bx,ax;
2039: double fx,fb,fa;
1.187 brouard 2040:
1.203 brouard 2041: #ifdef LINMINORIGINAL
2042: #else
2043: double scale=10., axs, xxs; /* Scale added for infinity */
2044: #endif
2045:
1.126 brouard 2046: ncom=n;
2047: pcom=vector(1,n);
2048: xicom=vector(1,n);
2049: nrfunc=func;
2050: for (j=1;j<=n;j++) {
2051: pcom[j]=p[j];
1.202 brouard 2052: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2053: }
1.187 brouard 2054:
1.203 brouard 2055: #ifdef LINMINORIGINAL
2056: xx=1.;
2057: #else
2058: axs=0.0;
2059: xxs=1.;
2060: do{
2061: xx= xxs;
2062: #endif
1.187 brouard 2063: ax=0.;
2064: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2065: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2066: /* 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)) */
2067: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2068: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2069: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2070: /* 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 2071: #ifdef LINMINORIGINAL
2072: #else
2073: if (fx != fx){
1.224 brouard 2074: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2075: printf("|");
2076: fprintf(ficlog,"|");
1.203 brouard 2077: #ifdef DEBUGLINMIN
1.224 brouard 2078: 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 2079: #endif
2080: }
1.224 brouard 2081: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2082: #endif
2083:
1.191 brouard 2084: #ifdef DEBUGLINMIN
2085: 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 2086: 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 2087: #endif
1.224 brouard 2088: #ifdef LINMINORIGINAL
2089: #else
2090: if(fb == fx){ /* Flat function in the direction */
2091: xmin=xx;
2092: *flat=1;
2093: }else{
2094: *flat=0;
2095: #endif
2096: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2097: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2098: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2099: /* fmin = f(p[j] + xmin * xi[j]) */
2100: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2101: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2102: #ifdef DEBUG
1.224 brouard 2103: 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);
2104: 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);
2105: #endif
2106: #ifdef LINMINORIGINAL
2107: #else
2108: }
1.126 brouard 2109: #endif
1.191 brouard 2110: #ifdef DEBUGLINMIN
2111: printf("linmin end ");
1.202 brouard 2112: fprintf(ficlog,"linmin end ");
1.191 brouard 2113: #endif
1.126 brouard 2114: for (j=1;j<=n;j++) {
1.203 brouard 2115: #ifdef LINMINORIGINAL
2116: xi[j] *= xmin;
2117: #else
2118: #ifdef DEBUGLINMIN
2119: if(xxs <1.0)
2120: printf(" before xi[%d]=%12.8f", j,xi[j]);
2121: #endif
2122: 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) */
2123: #ifdef DEBUGLINMIN
2124: if(xxs <1.0)
2125: 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 );
2126: #endif
2127: #endif
1.187 brouard 2128: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2129: }
1.191 brouard 2130: #ifdef DEBUGLINMIN
1.203 brouard 2131: printf("\n");
1.191 brouard 2132: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2133: 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 2134: for (j=1;j<=n;j++) {
1.202 brouard 2135: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2136: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2137: if(j % ncovmodel == 0){
1.191 brouard 2138: printf("\n");
1.202 brouard 2139: fprintf(ficlog,"\n");
2140: }
1.191 brouard 2141: }
1.203 brouard 2142: #else
1.191 brouard 2143: #endif
1.126 brouard 2144: free_vector(xicom,1,n);
2145: free_vector(pcom,1,n);
2146: }
2147:
2148:
2149: /*************** powell ************************/
1.162 brouard 2150: /*
2151: Minimization of a function func of n variables. Input consists of an initial starting point
2152: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2153: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2154: such that failure to decrease by more than this amount on one iteration signals doneness. On
2155: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2156: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2157: */
1.224 brouard 2158: #ifdef LINMINORIGINAL
2159: #else
2160: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2161: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2162: #endif
1.126 brouard 2163: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2164: double (*func)(double []))
2165: {
1.224 brouard 2166: #ifdef LINMINORIGINAL
2167: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2168: double (*func)(double []));
1.224 brouard 2169: #else
1.241 brouard 2170: void linmin(double p[], double xi[], int n, double *fret,
2171: double (*func)(double []),int *flat);
1.224 brouard 2172: #endif
1.239 brouard 2173: int i,ibig,j,jk,k;
1.126 brouard 2174: double del,t,*pt,*ptt,*xit;
1.181 brouard 2175: double directest;
1.126 brouard 2176: double fp,fptt;
2177: double *xits;
2178: int niterf, itmp;
1.224 brouard 2179: #ifdef LINMINORIGINAL
2180: #else
2181:
2182: flatdir=ivector(1,n);
2183: for (j=1;j<=n;j++) flatdir[j]=0;
2184: #endif
1.126 brouard 2185:
2186: pt=vector(1,n);
2187: ptt=vector(1,n);
2188: xit=vector(1,n);
2189: xits=vector(1,n);
2190: *fret=(*func)(p);
2191: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2192: rcurr_time = time(NULL);
1.126 brouard 2193: for (*iter=1;;++(*iter)) {
1.187 brouard 2194: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2195: ibig=0;
2196: del=0.0;
1.157 brouard 2197: rlast_time=rcurr_time;
2198: /* (void) gettimeofday(&curr_time,&tzp); */
2199: rcurr_time = time(NULL);
2200: curr_time = *localtime(&rcurr_time);
2201: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2202: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2203: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2204: for (i=1;i<=n;i++) {
1.126 brouard 2205: fprintf(ficrespow," %.12lf", p[i]);
2206: }
1.239 brouard 2207: fprintf(ficrespow,"\n");fflush(ficrespow);
2208: printf("\n#model= 1 + age ");
2209: fprintf(ficlog,"\n#model= 1 + age ");
2210: if(nagesqr==1){
1.241 brouard 2211: printf(" + age*age ");
2212: fprintf(ficlog," + age*age ");
1.239 brouard 2213: }
2214: for(j=1;j <=ncovmodel-2;j++){
2215: if(Typevar[j]==0) {
2216: printf(" + V%d ",Tvar[j]);
2217: fprintf(ficlog," + V%d ",Tvar[j]);
2218: }else if(Typevar[j]==1) {
2219: printf(" + V%d*age ",Tvar[j]);
2220: fprintf(ficlog," + V%d*age ",Tvar[j]);
2221: }else if(Typevar[j]==2) {
2222: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2223: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2224: }
2225: }
1.126 brouard 2226: printf("\n");
1.239 brouard 2227: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2228: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2229: fprintf(ficlog,"\n");
1.239 brouard 2230: for(i=1,jk=1; i <=nlstate; i++){
2231: for(k=1; k <=(nlstate+ndeath); k++){
2232: if (k != i) {
2233: printf("%d%d ",i,k);
2234: fprintf(ficlog,"%d%d ",i,k);
2235: for(j=1; j <=ncovmodel; j++){
2236: printf("%12.7f ",p[jk]);
2237: fprintf(ficlog,"%12.7f ",p[jk]);
2238: jk++;
2239: }
2240: printf("\n");
2241: fprintf(ficlog,"\n");
2242: }
2243: }
2244: }
1.241 brouard 2245: if(*iter <=3 && *iter >1){
1.157 brouard 2246: tml = *localtime(&rcurr_time);
2247: strcpy(strcurr,asctime(&tml));
2248: rforecast_time=rcurr_time;
1.126 brouard 2249: itmp = strlen(strcurr);
2250: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2251: strcurr[itmp-1]='\0';
1.162 brouard 2252: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2253: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2254: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2255: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2256: forecast_time = *localtime(&rforecast_time);
2257: strcpy(strfor,asctime(&forecast_time));
2258: itmp = strlen(strfor);
2259: if(strfor[itmp-1]=='\n')
2260: strfor[itmp-1]='\0';
2261: 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);
2262: 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 2263: }
2264: }
1.187 brouard 2265: for (i=1;i<=n;i++) { /* For each direction i */
2266: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2267: fptt=(*fret);
2268: #ifdef DEBUG
1.203 brouard 2269: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2270: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2271: #endif
1.203 brouard 2272: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2273: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2274: #ifdef LINMINORIGINAL
1.188 brouard 2275: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2276: #else
2277: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2278: flatdir[i]=flat; /* Function is vanishing in that direction i */
2279: #endif
2280: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2281: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2282: /* because that direction will be replaced unless the gain del is small */
2283: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2284: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2285: /* with the new direction. */
2286: del=fabs(fptt-(*fret));
2287: ibig=i;
1.126 brouard 2288: }
2289: #ifdef DEBUG
2290: printf("%d %.12e",i,(*fret));
2291: fprintf(ficlog,"%d %.12e",i,(*fret));
2292: for (j=1;j<=n;j++) {
1.224 brouard 2293: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2294: printf(" x(%d)=%.12e",j,xit[j]);
2295: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2296: }
2297: for(j=1;j<=n;j++) {
1.225 brouard 2298: printf(" p(%d)=%.12e",j,p[j]);
2299: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2300: }
2301: printf("\n");
2302: fprintf(ficlog,"\n");
2303: #endif
1.187 brouard 2304: } /* end loop on each direction i */
2305: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2306: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2307: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2308: for(j=1;j<=n;j++) {
1.225 brouard 2309: if(flatdir[j] >0){
2310: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2311: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2312: }
2313: /* printf("\n"); */
2314: /* fprintf(ficlog,"\n"); */
2315: }
1.243 brouard 2316: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2317: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2318: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2319: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2320: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2321: /* decreased of more than 3.84 */
2322: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2323: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2324: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2325:
1.188 brouard 2326: /* Starting the program with initial values given by a former maximization will simply change */
2327: /* the scales of the directions and the directions, because the are reset to canonical directions */
2328: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2329: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2330: #ifdef DEBUG
2331: int k[2],l;
2332: k[0]=1;
2333: k[1]=-1;
2334: printf("Max: %.12e",(*func)(p));
2335: fprintf(ficlog,"Max: %.12e",(*func)(p));
2336: for (j=1;j<=n;j++) {
2337: printf(" %.12e",p[j]);
2338: fprintf(ficlog," %.12e",p[j]);
2339: }
2340: printf("\n");
2341: fprintf(ficlog,"\n");
2342: for(l=0;l<=1;l++) {
2343: for (j=1;j<=n;j++) {
2344: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2345: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2346: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2347: }
2348: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2349: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2350: }
2351: #endif
2352:
1.224 brouard 2353: #ifdef LINMINORIGINAL
2354: #else
2355: free_ivector(flatdir,1,n);
2356: #endif
1.126 brouard 2357: free_vector(xit,1,n);
2358: free_vector(xits,1,n);
2359: free_vector(ptt,1,n);
2360: free_vector(pt,1,n);
2361: return;
1.192 brouard 2362: } /* enough precision */
1.240 brouard 2363: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2364: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2365: ptt[j]=2.0*p[j]-pt[j];
2366: xit[j]=p[j]-pt[j];
2367: pt[j]=p[j];
2368: }
1.181 brouard 2369: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2370: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2371: if (*iter <=4) {
1.225 brouard 2372: #else
2373: #endif
1.224 brouard 2374: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2375: #else
1.161 brouard 2376: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2377: #endif
1.162 brouard 2378: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2379: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2380: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2381: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2382: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2383: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2384: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2385: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2386: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2387: /* Even if f3 <f1, directest can be negative and t >0 */
2388: /* mu² and del² are equal when f3=f1 */
2389: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2390: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2391: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2392: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2393: #ifdef NRCORIGINAL
2394: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2395: #else
2396: 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 2397: t= t- del*SQR(fp-fptt);
1.183 brouard 2398: #endif
1.202 brouard 2399: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2400: #ifdef DEBUG
1.181 brouard 2401: 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);
2402: 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 2403: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2404: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2405: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2406: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2407: 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);
2408: 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);
2409: #endif
1.183 brouard 2410: #ifdef POWELLORIGINAL
2411: if (t < 0.0) { /* Then we use it for new direction */
2412: #else
1.182 brouard 2413: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2414: 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 2415: 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 2416: 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 2417: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2418: }
1.181 brouard 2419: if (directest < 0.0) { /* Then we use it for new direction */
2420: #endif
1.191 brouard 2421: #ifdef DEBUGLINMIN
1.234 brouard 2422: printf("Before linmin in direction P%d-P0\n",n);
2423: for (j=1;j<=n;j++) {
2424: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2425: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2426: if(j % ncovmodel == 0){
2427: printf("\n");
2428: fprintf(ficlog,"\n");
2429: }
2430: }
1.224 brouard 2431: #endif
2432: #ifdef LINMINORIGINAL
1.234 brouard 2433: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2434: #else
1.234 brouard 2435: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2436: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2437: #endif
1.234 brouard 2438:
1.191 brouard 2439: #ifdef DEBUGLINMIN
1.234 brouard 2440: for (j=1;j<=n;j++) {
2441: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2442: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2443: if(j % ncovmodel == 0){
2444: printf("\n");
2445: fprintf(ficlog,"\n");
2446: }
2447: }
1.224 brouard 2448: #endif
1.234 brouard 2449: for (j=1;j<=n;j++) {
2450: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2451: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2452: }
1.224 brouard 2453: #ifdef LINMINORIGINAL
2454: #else
1.234 brouard 2455: for (j=1, flatd=0;j<=n;j++) {
2456: if(flatdir[j]>0)
2457: flatd++;
2458: }
2459: if(flatd >0){
1.255 brouard 2460: printf("%d flat directions: ",flatd);
2461: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2462: for (j=1;j<=n;j++) {
2463: if(flatdir[j]>0){
2464: printf("%d ",j);
2465: fprintf(ficlog,"%d ",j);
2466: }
2467: }
2468: printf("\n");
2469: fprintf(ficlog,"\n");
2470: }
1.191 brouard 2471: #endif
1.234 brouard 2472: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2473: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2474:
1.126 brouard 2475: #ifdef DEBUG
1.234 brouard 2476: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2477: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2478: for(j=1;j<=n;j++){
2479: printf(" %lf",xit[j]);
2480: fprintf(ficlog," %lf",xit[j]);
2481: }
2482: printf("\n");
2483: fprintf(ficlog,"\n");
1.126 brouard 2484: #endif
1.192 brouard 2485: } /* end of t or directest negative */
1.224 brouard 2486: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2487: #else
1.234 brouard 2488: } /* end if (fptt < fp) */
1.192 brouard 2489: #endif
1.225 brouard 2490: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2491: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2492: #else
1.224 brouard 2493: #endif
1.234 brouard 2494: } /* loop iteration */
1.126 brouard 2495: }
1.234 brouard 2496:
1.126 brouard 2497: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2498:
1.235 brouard 2499: 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 2500: {
1.235 brouard 2501: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2502: (and selected quantitative values in nres)
2503: by left multiplying the unit
1.234 brouard 2504: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2505: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2506: /* Wx is row vector: population in state 1, population in state 2, population dead */
2507: /* or prevalence in state 1, prevalence in state 2, 0 */
2508: /* newm is the matrix after multiplications, its rows are identical at a factor */
2509: /* Initial matrix pimij */
2510: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2511: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2512: /* 0, 0 , 1} */
2513: /*
2514: * and after some iteration: */
2515: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2516: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2517: /* 0, 0 , 1} */
2518: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2519: /* {0.51571254859325999, 0.4842874514067399, */
2520: /* 0.51326036147820708, 0.48673963852179264} */
2521: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2522:
1.126 brouard 2523: int i, ii,j,k;
1.209 brouard 2524: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2525: /* double **matprod2(); */ /* test */
1.218 brouard 2526: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2527: double **newm;
1.209 brouard 2528: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2529: int ncvloop=0;
1.169 brouard 2530:
1.209 brouard 2531: min=vector(1,nlstate);
2532: max=vector(1,nlstate);
2533: meandiff=vector(1,nlstate);
2534:
1.218 brouard 2535: /* Starting with matrix unity */
1.126 brouard 2536: for (ii=1;ii<=nlstate+ndeath;ii++)
2537: for (j=1;j<=nlstate+ndeath;j++){
2538: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2539: }
1.169 brouard 2540:
2541: cov[1]=1.;
2542:
2543: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2544: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2545: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2546: ncvloop++;
1.126 brouard 2547: newm=savm;
2548: /* Covariates have to be included here again */
1.138 brouard 2549: cov[2]=agefin;
1.187 brouard 2550: if(nagesqr==1)
2551: cov[3]= agefin*agefin;;
1.234 brouard 2552: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2553: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2554: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2555: /* 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 2556: }
2557: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2558: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2559: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2560: /* 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 2561: }
1.237 brouard 2562: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2563: if(Dummy[Tvar[Tage[k]]]){
2564: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2565: } else{
1.235 brouard 2566: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2567: }
1.235 brouard 2568: /* 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 2569: }
1.237 brouard 2570: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2571: /* 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 2572: if(Dummy[Tvard[k][1]==0]){
2573: if(Dummy[Tvard[k][2]==0]){
2574: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2575: }else{
2576: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2577: }
2578: }else{
2579: if(Dummy[Tvard[k][2]==0]){
2580: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2581: }else{
2582: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2583: }
2584: }
1.234 brouard 2585: }
1.138 brouard 2586: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2587: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2588: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2589: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2590: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2591: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2592: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2593:
1.126 brouard 2594: savm=oldm;
2595: oldm=newm;
1.209 brouard 2596:
2597: for(j=1; j<=nlstate; j++){
2598: max[j]=0.;
2599: min[j]=1.;
2600: }
2601: for(i=1;i<=nlstate;i++){
2602: sumnew=0;
2603: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2604: for(j=1; j<=nlstate; j++){
2605: prlim[i][j]= newm[i][j]/(1-sumnew);
2606: max[j]=FMAX(max[j],prlim[i][j]);
2607: min[j]=FMIN(min[j],prlim[i][j]);
2608: }
2609: }
2610:
1.126 brouard 2611: maxmax=0.;
1.209 brouard 2612: for(j=1; j<=nlstate; j++){
2613: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2614: maxmax=FMAX(maxmax,meandiff[j]);
2615: /* 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 2616: } /* j loop */
1.203 brouard 2617: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2618: /* 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 2619: if(maxmax < ftolpl){
1.209 brouard 2620: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2621: free_vector(min,1,nlstate);
2622: free_vector(max,1,nlstate);
2623: free_vector(meandiff,1,nlstate);
1.126 brouard 2624: return prlim;
2625: }
1.169 brouard 2626: } /* age loop */
1.208 brouard 2627: /* After some age loop it doesn't converge */
1.209 brouard 2628: 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 2629: 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 2630: /* 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); */
2631: free_vector(min,1,nlstate);
2632: free_vector(max,1,nlstate);
2633: free_vector(meandiff,1,nlstate);
1.208 brouard 2634:
1.169 brouard 2635: return prlim; /* should not reach here */
1.126 brouard 2636: }
2637:
1.217 brouard 2638:
2639: /**** Back Prevalence limit (stable or period prevalence) ****************/
2640:
1.218 brouard 2641: /* 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) */
2642: /* 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 2643: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2644: {
1.264 brouard 2645: /* Computes the prevalence limit in each live state at age x and for covariate combination ij (<=2**cptcoveff) by left multiplying the unit
1.217 brouard 2646: matrix by transitions matrix until convergence is reached with precision ftolpl */
2647: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2648: /* Wx is row vector: population in state 1, population in state 2, population dead */
2649: /* or prevalence in state 1, prevalence in state 2, 0 */
2650: /* newm is the matrix after multiplications, its rows are identical at a factor */
2651: /* Initial matrix pimij */
2652: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2653: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2654: /* 0, 0 , 1} */
2655: /*
2656: * and after some iteration: */
2657: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2658: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2659: /* 0, 0 , 1} */
2660: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2661: /* {0.51571254859325999, 0.4842874514067399, */
2662: /* 0.51326036147820708, 0.48673963852179264} */
2663: /* If we start from prlim again, prlim tends to a constant matrix */
2664:
2665: int i, ii,j,k;
1.247 brouard 2666: int first=0;
1.217 brouard 2667: double *min, *max, *meandiff, maxmax,sumnew=0.;
2668: /* double **matprod2(); */ /* test */
2669: double **out, cov[NCOVMAX+1], **bmij();
2670: double **newm;
1.218 brouard 2671: double **dnewm, **doldm, **dsavm; /* for use */
2672: double **oldm, **savm; /* for use */
2673:
1.217 brouard 2674: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2675: int ncvloop=0;
2676:
2677: min=vector(1,nlstate);
2678: max=vector(1,nlstate);
2679: meandiff=vector(1,nlstate);
2680:
1.266 brouard 2681: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2682: oldm=oldms; savm=savms;
2683:
2684: /* Starting with matrix unity */
2685: for (ii=1;ii<=nlstate+ndeath;ii++)
2686: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2687: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2688: }
2689:
2690: cov[1]=1.;
2691:
2692: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2693: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2694: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2695: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2696: ncvloop++;
1.218 brouard 2697: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2698: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2699: /* Covariates have to be included here again */
2700: cov[2]=agefin;
2701: if(nagesqr==1)
2702: cov[3]= agefin*agefin;;
1.242 brouard 2703: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2704: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2705: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2706: /* printf("bprevalim Dummy agefin=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agefin,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
1.242 brouard 2707: }
2708: /* for (k=1; k<=cptcovn;k++) { */
2709: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2710: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2711: /* /\* 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])]); *\/ */
2712: /* } */
2713: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2714: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2715: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2716: /* 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]); */
2717: }
2718: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2719: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2720: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2721: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2722: for (k=1; k<=cptcovage;k++){ /* For product with age */
2723: if(Dummy[Tvar[Tage[k]]]){
2724: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2725: } else{
2726: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2727: }
2728: /* 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]); */
2729: }
2730: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2731: /* 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]); */
2732: if(Dummy[Tvard[k][1]==0]){
2733: if(Dummy[Tvard[k][2]==0]){
2734: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2735: }else{
2736: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2737: }
2738: }else{
2739: if(Dummy[Tvard[k][2]==0]){
2740: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2741: }else{
2742: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2743: }
2744: }
1.217 brouard 2745: }
2746:
2747: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2748: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2749: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2750: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2751: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2752: /* ij should be linked to the correct index of cov */
2753: /* age and covariate values ij are in 'cov', but we need to pass
2754: * ij for the observed prevalence at age and status and covariate
2755: * number: prevacurrent[(int)agefin][ii][ij]
2756: */
2757: /* 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 *\/ */
2758: /* 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 *\/ */
2759: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.268 brouard 2760: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2761: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2762: /* for(i=1; i<=nlstate+ndeath; i++) { */
2763: /* printf("%d newm= ",i); */
2764: /* for(j=1;j<=nlstate+ndeath;j++) { */
2765: /* printf("%f ",newm[i][j]); */
2766: /* } */
2767: /* printf("oldm * "); */
2768: /* for(j=1;j<=nlstate+ndeath;j++) { */
2769: /* printf("%f ",oldm[i][j]); */
2770: /* } */
1.268 brouard 2771: /* printf(" bmmij "); */
1.266 brouard 2772: /* for(j=1;j<=nlstate+ndeath;j++) { */
2773: /* printf("%f ",pmmij[i][j]); */
2774: /* } */
2775: /* printf("\n"); */
2776: /* } */
2777: /* } */
1.217 brouard 2778: savm=oldm;
2779: oldm=newm;
1.266 brouard 2780:
1.217 brouard 2781: for(j=1; j<=nlstate; j++){
2782: max[j]=0.;
2783: min[j]=1.;
2784: }
2785: for(j=1; j<=nlstate; j++){
2786: for(i=1;i<=nlstate;i++){
1.234 brouard 2787: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2788: bprlim[i][j]= newm[i][j];
2789: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2790: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2791: }
2792: }
1.218 brouard 2793:
1.217 brouard 2794: maxmax=0.;
2795: for(i=1; i<=nlstate; i++){
2796: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2797: maxmax=FMAX(maxmax,meandiff[i]);
2798: /* 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); */
1.268 brouard 2799: } /* i loop */
1.217 brouard 2800: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2801: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2802: if(maxmax < ftolpl){
1.220 brouard 2803: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2804: free_vector(min,1,nlstate);
2805: free_vector(max,1,nlstate);
2806: free_vector(meandiff,1,nlstate);
2807: return bprlim;
2808: }
2809: } /* age loop */
2810: /* After some age loop it doesn't converge */
1.247 brouard 2811: if(first){
2812: first=1;
2813: 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\
2814: 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);
2815: }
2816: 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 2817: 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);
2818: /* 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); */
2819: free_vector(min,1,nlstate);
2820: free_vector(max,1,nlstate);
2821: free_vector(meandiff,1,nlstate);
2822:
2823: return bprlim; /* should not reach here */
2824: }
2825:
1.126 brouard 2826: /*************** transition probabilities ***************/
2827:
2828: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2829: {
1.138 brouard 2830: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2831: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2832: model to the ncovmodel covariates (including constant and age).
2833: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2834: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2835: ncth covariate in the global vector x is given by the formula:
2836: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2837: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2838: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2839: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2840: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2841: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2842: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2843: */
2844: double s1, lnpijopii;
1.126 brouard 2845: /*double t34;*/
1.164 brouard 2846: int i,j, nc, ii, jj;
1.126 brouard 2847:
1.223 brouard 2848: for(i=1; i<= nlstate; i++){
2849: for(j=1; j<i;j++){
2850: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2851: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2852: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2853: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2854: }
2855: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2856: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2857: }
2858: for(j=i+1; j<=nlstate+ndeath;j++){
2859: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2860: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2861: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2862: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2863: }
2864: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2865: }
2866: }
1.218 brouard 2867:
1.223 brouard 2868: for(i=1; i<= nlstate; i++){
2869: s1=0;
2870: for(j=1; j<i; j++){
2871: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2872: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2873: }
2874: for(j=i+1; j<=nlstate+ndeath; j++){
2875: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2876: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2877: }
2878: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2879: ps[i][i]=1./(s1+1.);
2880: /* Computing other pijs */
2881: for(j=1; j<i; j++)
2882: ps[i][j]= exp(ps[i][j])*ps[i][i];
2883: for(j=i+1; j<=nlstate+ndeath; j++)
2884: ps[i][j]= exp(ps[i][j])*ps[i][i];
2885: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2886: } /* end i */
1.218 brouard 2887:
1.223 brouard 2888: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2889: for(jj=1; jj<= nlstate+ndeath; jj++){
2890: ps[ii][jj]=0;
2891: ps[ii][ii]=1;
2892: }
2893: }
1.218 brouard 2894:
2895:
1.223 brouard 2896: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2897: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2898: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2899: /* } */
2900: /* printf("\n "); */
2901: /* } */
2902: /* printf("\n ");printf("%lf ",cov[2]);*/
2903: /*
2904: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2905: goto end;*/
1.266 brouard 2906: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2907: }
2908:
1.218 brouard 2909: /*************** backward transition probabilities ***************/
2910:
2911: /* 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 ) */
2912: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2913: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2914: {
1.266 brouard 2915: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2916: * Call to pmij(cov and x), call to cross prevalence, sums and inverses, left multiply, and returns in **ps as well as **bmij.
1.222 brouard 2917: */
1.218 brouard 2918: int i, ii, j,k;
1.222 brouard 2919:
2920: double **out, **pmij();
2921: double sumnew=0.;
1.218 brouard 2922: double agefin;
1.268 brouard 2923: double k3=0.; /* constant of the w_x diagonal matrixe (in order for B to sum to 1 even for death state) */
1.222 brouard 2924: double **dnewm, **dsavm, **doldm;
2925: double **bbmij;
2926:
1.218 brouard 2927: doldm=ddoldms; /* global pointers */
1.222 brouard 2928: dnewm=ddnewms;
2929: dsavm=ddsavms;
2930:
2931: agefin=cov[2];
1.268 brouard 2932: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2933: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2934: the observed prevalence (with this covariate ij) at beginning of transition */
2935: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2936:
2937: /* P_x */
1.266 brouard 2938: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2939: /* outputs pmmij which is a stochastic matrix in row */
2940:
2941: /* Diag(w_x) */
2942: /* Problem with prevacurrent which can be zero */
2943: sumnew=0.;
1.269 brouard 2944: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2945: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2946: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2947: sumnew+=prevacurrent[(int)agefin][ii][ij];
2948: }
2949: if(sumnew >0.01){ /* At least some value in the prevalence */
2950: for (ii=1;ii<=nlstate+ndeath;ii++){
2951: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2952: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2953: }
2954: }else{
2955: for (ii=1;ii<=nlstate+ndeath;ii++){
2956: for (j=1;j<=nlstate+ndeath;j++)
2957: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2958: }
2959: /* if(sumnew <0.9){ */
2960: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2961: /* } */
2962: }
2963: k3=0.0; /* We put the last diagonal to 0 */
2964: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2965: doldm[ii][ii]= k3;
2966: }
2967: /* End doldm, At the end doldm is diag[(w_i)] */
2968:
2969: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2970: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2971:
2972: /* Diag(Sum_i w^i_x p^ij_x */
2973: /* w1 p11 + w2 p21 only on live states N1./N..*N11/N1. + N2./N..*N21/N2.=(N11+N21)/N..=N.1/N.. */
1.222 brouard 2974: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2975: sumnew=0.;
1.222 brouard 2976: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2977: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2978: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2979: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2980: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2981: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2982: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2983: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2984: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2985: /* }else */
1.268 brouard 2986: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2987: } /*End ii */
2988: } /* End j, At the end dsavm is diag[1/(w_1p1i+w_2 p2i)] for ALL states even if the sum is only for live states */
2989:
2990: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
2991: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 2992: /* end bmij */
1.266 brouard 2993: return ps; /*pointer is unchanged */
1.218 brouard 2994: }
1.217 brouard 2995: /*************** transition probabilities ***************/
2996:
1.218 brouard 2997: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2998: {
2999: /* According to parameters values stored in x and the covariate's values stored in cov,
3000: computes the probability to be observed in state j being in state i by appying the
3001: model to the ncovmodel covariates (including constant and age).
3002: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3003: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3004: ncth covariate in the global vector x is given by the formula:
3005: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3006: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3007: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3008: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3009: Outputs ps[i][j] the probability to be observed in j being in j according to
3010: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3011: */
3012: double s1, lnpijopii;
3013: /*double t34;*/
3014: int i,j, nc, ii, jj;
3015:
1.234 brouard 3016: for(i=1; i<= nlstate; i++){
3017: for(j=1; j<i;j++){
3018: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3019: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3020: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3021: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3022: }
3023: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3024: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3025: }
3026: for(j=i+1; j<=nlstate+ndeath;j++){
3027: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3028: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3029: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3030: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3031: }
3032: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3033: }
3034: }
3035:
3036: for(i=1; i<= nlstate; i++){
3037: s1=0;
3038: for(j=1; j<i; j++){
3039: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3040: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3041: }
3042: for(j=i+1; j<=nlstate+ndeath; j++){
3043: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3044: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3045: }
3046: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3047: ps[i][i]=1./(s1+1.);
3048: /* Computing other pijs */
3049: for(j=1; j<i; j++)
3050: ps[i][j]= exp(ps[i][j])*ps[i][i];
3051: for(j=i+1; j<=nlstate+ndeath; j++)
3052: ps[i][j]= exp(ps[i][j])*ps[i][i];
3053: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3054: } /* end i */
3055:
3056: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3057: for(jj=1; jj<= nlstate+ndeath; jj++){
3058: ps[ii][jj]=0;
3059: ps[ii][ii]=1;
3060: }
3061: }
3062: /* Added for backcast */ /* Transposed matrix too */
3063: for(jj=1; jj<= nlstate+ndeath; jj++){
3064: s1=0.;
3065: for(ii=1; ii<= nlstate+ndeath; ii++){
3066: s1+=ps[ii][jj];
3067: }
3068: for(ii=1; ii<= nlstate; ii++){
3069: ps[ii][jj]=ps[ii][jj]/s1;
3070: }
3071: }
3072: /* Transposition */
3073: for(jj=1; jj<= nlstate+ndeath; jj++){
3074: for(ii=jj; ii<= nlstate+ndeath; ii++){
3075: s1=ps[ii][jj];
3076: ps[ii][jj]=ps[jj][ii];
3077: ps[jj][ii]=s1;
3078: }
3079: }
3080: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3081: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3082: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3083: /* } */
3084: /* printf("\n "); */
3085: /* } */
3086: /* printf("\n ");printf("%lf ",cov[2]);*/
3087: /*
3088: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3089: goto end;*/
3090: return ps;
1.217 brouard 3091: }
3092:
3093:
1.126 brouard 3094: /**************** Product of 2 matrices ******************/
3095:
1.145 brouard 3096: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3097: {
3098: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3099: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3100: /* in, b, out are matrice of pointers which should have been initialized
3101: before: only the contents of out is modified. The function returns
3102: a pointer to pointers identical to out */
1.145 brouard 3103: int i, j, k;
1.126 brouard 3104: for(i=nrl; i<= nrh; i++)
1.145 brouard 3105: for(k=ncolol; k<=ncoloh; k++){
3106: out[i][k]=0.;
3107: for(j=ncl; j<=nch; j++)
3108: out[i][k] +=in[i][j]*b[j][k];
3109: }
1.126 brouard 3110: return out;
3111: }
3112:
3113:
3114: /************* Higher Matrix Product ***************/
3115:
1.235 brouard 3116: 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 3117: {
1.218 brouard 3118: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3119: 'nhstepm*hstepm*stepm' months (i.e. until
3120: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3121: nhstepm*hstepm matrices.
3122: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3123: (typically every 2 years instead of every month which is too big
3124: for the memory).
3125: Model is determined by parameters x and covariates have to be
3126: included manually here.
3127:
3128: */
3129:
3130: int i, j, d, h, k;
1.131 brouard 3131: double **out, cov[NCOVMAX+1];
1.126 brouard 3132: double **newm;
1.187 brouard 3133: double agexact;
1.214 brouard 3134: double agebegin, ageend;
1.126 brouard 3135:
3136: /* Hstepm could be zero and should return the unit matrix */
3137: for (i=1;i<=nlstate+ndeath;i++)
3138: for (j=1;j<=nlstate+ndeath;j++){
3139: oldm[i][j]=(i==j ? 1.0 : 0.0);
3140: po[i][j][0]=(i==j ? 1.0 : 0.0);
3141: }
3142: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3143: for(h=1; h <=nhstepm; h++){
3144: for(d=1; d <=hstepm; d++){
3145: newm=savm;
3146: /* Covariates have to be included here again */
3147: cov[1]=1.;
1.214 brouard 3148: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3149: cov[2]=agexact;
3150: if(nagesqr==1)
1.227 brouard 3151: cov[3]= agexact*agexact;
1.235 brouard 3152: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3153: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3154: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3155: /* 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)); */
3156: }
3157: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3158: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3159: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3160: /* 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]); */
3161: }
3162: for (k=1; k<=cptcovage;k++){
3163: if(Dummy[Tvar[Tage[k]]]){
3164: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3165: } else{
3166: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3167: }
3168: /* 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]); */
3169: }
3170: for (k=1; k<=cptcovprod;k++){ /* */
3171: /* 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]); */
3172: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3173: }
3174: /* for (k=1; k<=cptcovn;k++) */
3175: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3176: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3177: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3178: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3179: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3180:
3181:
1.126 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: /* right multiplication of oldm by the current matrix */
1.126 brouard 3185: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3186: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3187: /* if((int)age == 70){ */
3188: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3189: /* for(i=1; i<=nlstate+ndeath; i++) { */
3190: /* printf("%d pmmij ",i); */
3191: /* for(j=1;j<=nlstate+ndeath;j++) { */
3192: /* printf("%f ",pmmij[i][j]); */
3193: /* } */
3194: /* printf(" oldm "); */
3195: /* for(j=1;j<=nlstate+ndeath;j++) { */
3196: /* printf("%f ",oldm[i][j]); */
3197: /* } */
3198: /* printf("\n"); */
3199: /* } */
3200: /* } */
1.126 brouard 3201: savm=oldm;
3202: oldm=newm;
3203: }
3204: for(i=1; i<=nlstate+ndeath; i++)
3205: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3206: po[i][j][h]=newm[i][j];
3207: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3208: }
1.128 brouard 3209: /*printf("h=%d ",h);*/
1.126 brouard 3210: } /* end h */
1.267 brouard 3211: /* printf("\n H=%d \n",h); */
1.126 brouard 3212: return po;
3213: }
3214:
1.217 brouard 3215: /************* Higher Back Matrix Product ***************/
1.218 brouard 3216: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.267 brouard 3217: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij, int nres )
1.217 brouard 3218: {
1.266 brouard 3219: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3220: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3221: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3222: nhstepm*hstepm matrices.
3223: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3224: (typically every 2 years instead of every month which is too big
1.217 brouard 3225: for the memory).
1.218 brouard 3226: Model is determined by parameters x and covariates have to be
1.266 brouard 3227: included manually here. Then we use a call to bmij(x and cov)
3228: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3229: */
1.217 brouard 3230:
3231: int i, j, d, h, k;
1.266 brouard 3232: double **out, cov[NCOVMAX+1], **bmij();
3233: double **newm, ***newmm;
1.217 brouard 3234: double agexact;
3235: double agebegin, ageend;
1.222 brouard 3236: double **oldm, **savm;
1.217 brouard 3237:
1.266 brouard 3238: newmm=po; /* To be saved */
3239: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3240: /* Hstepm could be zero and should return the unit matrix */
3241: for (i=1;i<=nlstate+ndeath;i++)
3242: for (j=1;j<=nlstate+ndeath;j++){
3243: oldm[i][j]=(i==j ? 1.0 : 0.0);
3244: po[i][j][0]=(i==j ? 1.0 : 0.0);
3245: }
3246: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3247: for(h=1; h <=nhstepm; h++){
3248: for(d=1; d <=hstepm; d++){
3249: newm=savm;
3250: /* Covariates have to be included here again */
3251: cov[1]=1.;
1.271 brouard 3252: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3253: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3254: cov[2]=agexact;
3255: if(nagesqr==1)
1.222 brouard 3256: cov[3]= agexact*agexact;
1.266 brouard 3257: for (k=1; k<=cptcovn;k++){
3258: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3259: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3260: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3261: /* printf("hbxij Dummy agexact=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agexact,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
3262: }
1.267 brouard 3263: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3264: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3265: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3266: /* 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]); */
3267: }
3268: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3269: if(Dummy[Tvar[Tage[k]]]){
3270: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3271: } else{
3272: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3273: }
3274: /* printf("hBxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); */
3275: }
3276: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3277: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3278: }
1.217 brouard 3279: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3280: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3281:
1.218 brouard 3282: /* Careful transposed matrix */
1.266 brouard 3283: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3284: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3285: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3286: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3287: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3288: /* if((int)age == 70){ */
3289: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3290: /* for(i=1; i<=nlstate+ndeath; i++) { */
3291: /* printf("%d pmmij ",i); */
3292: /* for(j=1;j<=nlstate+ndeath;j++) { */
3293: /* printf("%f ",pmmij[i][j]); */
3294: /* } */
3295: /* printf(" oldm "); */
3296: /* for(j=1;j<=nlstate+ndeath;j++) { */
3297: /* printf("%f ",oldm[i][j]); */
3298: /* } */
3299: /* printf("\n"); */
3300: /* } */
3301: /* } */
3302: savm=oldm;
3303: oldm=newm;
3304: }
3305: for(i=1; i<=nlstate+ndeath; i++)
3306: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3307: po[i][j][h]=newm[i][j];
1.268 brouard 3308: /* if(h==nhstepm) */
3309: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3310: }
1.268 brouard 3311: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3312: } /* end h */
1.268 brouard 3313: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3314: return po;
3315: }
3316:
3317:
1.162 brouard 3318: #ifdef NLOPT
3319: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3320: double fret;
3321: double *xt;
3322: int j;
3323: myfunc_data *d2 = (myfunc_data *) pd;
3324: /* xt = (p1-1); */
3325: xt=vector(1,n);
3326: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3327:
3328: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3329: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3330: printf("Function = %.12lf ",fret);
3331: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3332: printf("\n");
3333: free_vector(xt,1,n);
3334: return fret;
3335: }
3336: #endif
1.126 brouard 3337:
3338: /*************** log-likelihood *************/
3339: double func( double *x)
3340: {
1.226 brouard 3341: int i, ii, j, k, mi, d, kk;
3342: int ioffset=0;
3343: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3344: double **out;
3345: double lli; /* Individual log likelihood */
3346: int s1, s2;
1.228 brouard 3347: 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 3348: double bbh, survp;
3349: long ipmx;
3350: double agexact;
3351: /*extern weight */
3352: /* We are differentiating ll according to initial status */
3353: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3354: /*for(i=1;i<imx;i++)
3355: printf(" %d\n",s[4][i]);
3356: */
1.162 brouard 3357:
1.226 brouard 3358: ++countcallfunc;
1.162 brouard 3359:
1.226 brouard 3360: cov[1]=1.;
1.126 brouard 3361:
1.226 brouard 3362: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3363: ioffset=0;
1.226 brouard 3364: if(mle==1){
3365: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3366: /* Computes the values of the ncovmodel covariates of the model
3367: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3368: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3369: to be observed in j being in i according to the model.
3370: */
1.243 brouard 3371: ioffset=2+nagesqr ;
1.233 brouard 3372: /* Fixed */
1.234 brouard 3373: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3374: 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)*/
3375: }
1.226 brouard 3376: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3377: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3378: has been calculated etc */
3379: /* For an individual i, wav[i] gives the number of effective waves */
3380: /* We compute the contribution to Likelihood of each effective transition
3381: mw[mi][i] is real wave of the mi th effectve wave */
3382: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3383: s2=s[mw[mi+1][i]][i];
3384: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3385: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3386: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3387: */
3388: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3389: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3390: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3391: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3392: }
3393: for (ii=1;ii<=nlstate+ndeath;ii++)
3394: for (j=1;j<=nlstate+ndeath;j++){
3395: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3396: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3397: }
3398: for(d=0; d<dh[mi][i]; d++){
3399: newm=savm;
3400: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3401: cov[2]=agexact;
3402: if(nagesqr==1)
3403: cov[3]= agexact*agexact; /* Should be changed here */
3404: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3405: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3406: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3407: else
3408: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3409: }
3410: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3411: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3412: savm=oldm;
3413: oldm=newm;
3414: } /* end mult */
3415:
3416: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3417: /* But now since version 0.9 we anticipate for bias at large stepm.
3418: * If stepm is larger than one month (smallest stepm) and if the exact delay
3419: * (in months) between two waves is not a multiple of stepm, we rounded to
3420: * the nearest (and in case of equal distance, to the lowest) interval but now
3421: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3422: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3423: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3424: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3425: * -stepm/2 to stepm/2 .
3426: * For stepm=1 the results are the same as for previous versions of Imach.
3427: * For stepm > 1 the results are less biased than in previous versions.
3428: */
1.234 brouard 3429: s1=s[mw[mi][i]][i];
3430: s2=s[mw[mi+1][i]][i];
3431: bbh=(double)bh[mi][i]/(double)stepm;
3432: /* bias bh is positive if real duration
3433: * is higher than the multiple of stepm and negative otherwise.
3434: */
3435: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3436: if( s2 > nlstate){
3437: /* i.e. if s2 is a death state and if the date of death is known
3438: then the contribution to the likelihood is the probability to
3439: die between last step unit time and current step unit time,
3440: which is also equal to probability to die before dh
3441: minus probability to die before dh-stepm .
3442: In version up to 0.92 likelihood was computed
3443: as if date of death was unknown. Death was treated as any other
3444: health state: the date of the interview describes the actual state
3445: and not the date of a change in health state. The former idea was
3446: to consider that at each interview the state was recorded
3447: (healthy, disable or death) and IMaCh was corrected; but when we
3448: introduced the exact date of death then we should have modified
3449: the contribution of an exact death to the likelihood. This new
3450: contribution is smaller and very dependent of the step unit
3451: stepm. It is no more the probability to die between last interview
3452: and month of death but the probability to survive from last
3453: interview up to one month before death multiplied by the
3454: probability to die within a month. Thanks to Chris
3455: Jackson for correcting this bug. Former versions increased
3456: mortality artificially. The bad side is that we add another loop
3457: which slows down the processing. The difference can be up to 10%
3458: lower mortality.
3459: */
3460: /* If, at the beginning of the maximization mostly, the
3461: cumulative probability or probability to be dead is
3462: constant (ie = 1) over time d, the difference is equal to
3463: 0. out[s1][3] = savm[s1][3]: probability, being at state
3464: s1 at precedent wave, to be dead a month before current
3465: wave is equal to probability, being at state s1 at
3466: precedent wave, to be dead at mont of the current
3467: wave. Then the observed probability (that this person died)
3468: is null according to current estimated parameter. In fact,
3469: it should be very low but not zero otherwise the log go to
3470: infinity.
3471: */
1.183 brouard 3472: /* #ifdef INFINITYORIGINAL */
3473: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3474: /* #else */
3475: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3476: /* lli=log(mytinydouble); */
3477: /* else */
3478: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3479: /* #endif */
1.226 brouard 3480: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3481:
1.226 brouard 3482: } else if ( s2==-1 ) { /* alive */
3483: for (j=1,survp=0. ; j<=nlstate; j++)
3484: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3485: /*survp += out[s1][j]; */
3486: lli= log(survp);
3487: }
3488: else if (s2==-4) {
3489: for (j=3,survp=0. ; j<=nlstate; j++)
3490: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3491: lli= log(survp);
3492: }
3493: else if (s2==-5) {
3494: for (j=1,survp=0. ; j<=2; j++)
3495: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3496: lli= log(survp);
3497: }
3498: else{
3499: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3500: /* 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 */
3501: }
3502: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3503: /*if(lli ==000.0)*/
3504: /*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); */
3505: ipmx +=1;
3506: sw += weight[i];
3507: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3508: /* if (lli < log(mytinydouble)){ */
3509: /* 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); */
3510: /* 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]); */
3511: /* } */
3512: } /* end of wave */
3513: } /* end of individual */
3514: } else if(mle==2){
3515: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3516: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3517: for(mi=1; mi<= wav[i]-1; mi++){
3518: for (ii=1;ii<=nlstate+ndeath;ii++)
3519: for (j=1;j<=nlstate+ndeath;j++){
3520: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3521: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3522: }
3523: for(d=0; d<=dh[mi][i]; d++){
3524: newm=savm;
3525: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3526: cov[2]=agexact;
3527: if(nagesqr==1)
3528: cov[3]= agexact*agexact;
3529: for (kk=1; kk<=cptcovage;kk++) {
3530: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3531: }
3532: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3533: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3534: savm=oldm;
3535: oldm=newm;
3536: } /* end mult */
3537:
3538: s1=s[mw[mi][i]][i];
3539: s2=s[mw[mi+1][i]][i];
3540: bbh=(double)bh[mi][i]/(double)stepm;
3541: 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 */
3542: ipmx +=1;
3543: sw += weight[i];
3544: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3545: } /* end of wave */
3546: } /* end of individual */
3547: } else if(mle==3){ /* exponential inter-extrapolation */
3548: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3549: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3550: for(mi=1; mi<= wav[i]-1; mi++){
3551: for (ii=1;ii<=nlstate+ndeath;ii++)
3552: for (j=1;j<=nlstate+ndeath;j++){
3553: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3554: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3555: }
3556: for(d=0; d<dh[mi][i]; d++){
3557: newm=savm;
3558: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3559: cov[2]=agexact;
3560: if(nagesqr==1)
3561: cov[3]= agexact*agexact;
3562: for (kk=1; kk<=cptcovage;kk++) {
3563: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3564: }
3565: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3566: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3567: savm=oldm;
3568: oldm=newm;
3569: } /* end mult */
3570:
3571: s1=s[mw[mi][i]][i];
3572: s2=s[mw[mi+1][i]][i];
3573: bbh=(double)bh[mi][i]/(double)stepm;
3574: 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 */
3575: ipmx +=1;
3576: sw += weight[i];
3577: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3578: } /* end of wave */
3579: } /* end of individual */
3580: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3581: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3582: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3583: for(mi=1; mi<= wav[i]-1; mi++){
3584: for (ii=1;ii<=nlstate+ndeath;ii++)
3585: for (j=1;j<=nlstate+ndeath;j++){
3586: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3587: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3588: }
3589: for(d=0; d<dh[mi][i]; d++){
3590: newm=savm;
3591: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3592: cov[2]=agexact;
3593: if(nagesqr==1)
3594: cov[3]= agexact*agexact;
3595: for (kk=1; kk<=cptcovage;kk++) {
3596: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3597: }
1.126 brouard 3598:
1.226 brouard 3599: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3600: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3601: savm=oldm;
3602: oldm=newm;
3603: } /* end mult */
3604:
3605: s1=s[mw[mi][i]][i];
3606: s2=s[mw[mi+1][i]][i];
3607: if( s2 > nlstate){
3608: lli=log(out[s1][s2] - savm[s1][s2]);
3609: } else if ( s2==-1 ) { /* alive */
3610: for (j=1,survp=0. ; j<=nlstate; j++)
3611: survp += out[s1][j];
3612: lli= log(survp);
3613: }else{
3614: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3615: }
3616: ipmx +=1;
3617: sw += weight[i];
3618: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3619: /* 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 3620: } /* end of wave */
3621: } /* end of individual */
3622: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3623: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3624: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3625: for(mi=1; mi<= wav[i]-1; mi++){
3626: for (ii=1;ii<=nlstate+ndeath;ii++)
3627: for (j=1;j<=nlstate+ndeath;j++){
3628: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3629: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3630: }
3631: for(d=0; d<dh[mi][i]; d++){
3632: newm=savm;
3633: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3634: cov[2]=agexact;
3635: if(nagesqr==1)
3636: cov[3]= agexact*agexact;
3637: for (kk=1; kk<=cptcovage;kk++) {
3638: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3639: }
1.126 brouard 3640:
1.226 brouard 3641: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3642: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3643: savm=oldm;
3644: oldm=newm;
3645: } /* end mult */
3646:
3647: s1=s[mw[mi][i]][i];
3648: s2=s[mw[mi+1][i]][i];
3649: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3650: ipmx +=1;
3651: sw += weight[i];
3652: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3653: /*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]);*/
3654: } /* end of wave */
3655: } /* end of individual */
3656: } /* End of if */
3657: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3658: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3659: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3660: return -l;
1.126 brouard 3661: }
3662:
3663: /*************** log-likelihood *************/
3664: double funcone( double *x)
3665: {
1.228 brouard 3666: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3667: int i, ii, j, k, mi, d, kk;
1.228 brouard 3668: int ioffset=0;
1.131 brouard 3669: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3670: double **out;
3671: double lli; /* Individual log likelihood */
3672: double llt;
3673: int s1, s2;
1.228 brouard 3674: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3675:
1.126 brouard 3676: double bbh, survp;
1.187 brouard 3677: double agexact;
1.214 brouard 3678: double agebegin, ageend;
1.126 brouard 3679: /*extern weight */
3680: /* We are differentiating ll according to initial status */
3681: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3682: /*for(i=1;i<imx;i++)
3683: printf(" %d\n",s[4][i]);
3684: */
3685: cov[1]=1.;
3686:
3687: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3688: ioffset=0;
3689: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3690: /* ioffset=2+nagesqr+cptcovage; */
3691: ioffset=2+nagesqr;
1.232 brouard 3692: /* Fixed */
1.224 brouard 3693: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3694: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3695: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3696: 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)*/
3697: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3698: /* cov[2+6]=covar[Tvar[6]][i]; */
3699: /* cov[2+6]=covar[2][i]; V2 */
3700: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3701: /* cov[2+7]=covar[Tvar[7]][i]; */
3702: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3703: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3704: /* cov[2+9]=covar[Tvar[9]][i]; */
3705: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3706: }
1.232 brouard 3707: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3708: /* 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?)*\/ */
3709: /* } */
1.231 brouard 3710: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3711: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3712: /* } */
1.225 brouard 3713:
1.233 brouard 3714:
3715: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3716: /* Wave varying (but not age varying) */
3717: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3718: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3719: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3720: }
1.232 brouard 3721: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3722: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3723: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3724: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3725: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3726: /* 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 3727: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3728: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3729: /* /\* 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]); *\/ */
3730: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3731: /* } */
1.126 brouard 3732: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3733: for (j=1;j<=nlstate+ndeath;j++){
3734: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3735: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3736: }
1.214 brouard 3737:
3738: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3739: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3740: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3741: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3742: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3743: and mw[mi+1][i]. dh depends on stepm.*/
3744: newm=savm;
1.247 brouard 3745: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3746: cov[2]=agexact;
3747: if(nagesqr==1)
3748: cov[3]= agexact*agexact;
3749: for (kk=1; kk<=cptcovage;kk++) {
3750: if(!FixedV[Tvar[Tage[kk]]])
3751: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3752: else
3753: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3754: }
3755: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3756: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3757: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3758: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3759: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3760: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3761: savm=oldm;
3762: oldm=newm;
1.126 brouard 3763: } /* end mult */
3764:
3765: s1=s[mw[mi][i]][i];
3766: s2=s[mw[mi+1][i]][i];
1.217 brouard 3767: /* if(s2==-1){ */
1.268 brouard 3768: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3769: /* /\* exit(1); *\/ */
3770: /* } */
1.126 brouard 3771: bbh=(double)bh[mi][i]/(double)stepm;
3772: /* bias is positive if real duration
3773: * is higher than the multiple of stepm and negative otherwise.
3774: */
3775: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3776: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3777: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3778: for (j=1,survp=0. ; j<=nlstate; j++)
3779: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3780: lli= log(survp);
1.126 brouard 3781: }else if (mle==1){
1.242 brouard 3782: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3783: } else if(mle==2){
1.242 brouard 3784: 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 3785: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3786: 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 3787: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3788: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3789: } else{ /* mle=0 back to 1 */
1.242 brouard 3790: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3791: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3792: } /* End of if */
3793: ipmx +=1;
3794: sw += weight[i];
3795: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3796: /*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 3797: if(globpr){
1.246 brouard 3798: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3799: %11.6f %11.6f %11.6f ", \
1.242 brouard 3800: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
1.268 brouard 3801: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3802: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3803: llt +=ll[k]*gipmx/gsw;
3804: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3805: }
3806: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3807: }
1.232 brouard 3808: } /* end of wave */
3809: } /* end of individual */
3810: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3811: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3812: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3813: if(globpr==0){ /* First time we count the contributions and weights */
3814: gipmx=ipmx;
3815: gsw=sw;
3816: }
3817: return -l;
1.126 brouard 3818: }
3819:
3820:
3821: /*************** function likelione ***********/
3822: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3823: {
3824: /* This routine should help understanding what is done with
3825: the selection of individuals/waves and
3826: to check the exact contribution to the likelihood.
3827: Plotting could be done.
3828: */
3829: int k;
3830:
3831: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3832: strcpy(fileresilk,"ILK_");
1.202 brouard 3833: strcat(fileresilk,fileresu);
1.126 brouard 3834: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3835: printf("Problem with resultfile: %s\n", fileresilk);
3836: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3837: }
1.214 brouard 3838: 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");
3839: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3840: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3841: for(k=1; k<=nlstate; k++)
3842: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3843: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3844: }
3845:
3846: *fretone=(*funcone)(p);
3847: if(*globpri !=0){
3848: fclose(ficresilk);
1.205 brouard 3849: if (mle ==0)
3850: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3851: else if(mle >=1)
3852: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3853: 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 3854:
1.208 brouard 3855:
3856: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3857: 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 3858: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3859: }
1.207 brouard 3860: 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 3861: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3862: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3863: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3864: fflush(fichtm);
1.205 brouard 3865: }
1.126 brouard 3866: return;
3867: }
3868:
3869:
3870: /*********** Maximum Likelihood Estimation ***************/
3871:
3872: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3873: {
1.165 brouard 3874: int i,j, iter=0;
1.126 brouard 3875: double **xi;
3876: double fret;
3877: double fretone; /* Only one call to likelihood */
3878: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3879:
3880: #ifdef NLOPT
3881: int creturn;
3882: nlopt_opt opt;
3883: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3884: double *lb;
3885: double minf; /* the minimum objective value, upon return */
3886: double * p1; /* Shifted parameters from 0 instead of 1 */
3887: myfunc_data dinst, *d = &dinst;
3888: #endif
3889:
3890:
1.126 brouard 3891: xi=matrix(1,npar,1,npar);
3892: for (i=1;i<=npar;i++)
3893: for (j=1;j<=npar;j++)
3894: xi[i][j]=(i==j ? 1.0 : 0.0);
3895: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3896: strcpy(filerespow,"POW_");
1.126 brouard 3897: strcat(filerespow,fileres);
3898: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3899: printf("Problem with resultfile: %s\n", filerespow);
3900: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3901: }
3902: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3903: for (i=1;i<=nlstate;i++)
3904: for(j=1;j<=nlstate+ndeath;j++)
3905: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3906: fprintf(ficrespow,"\n");
1.162 brouard 3907: #ifdef POWELL
1.126 brouard 3908: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3909: #endif
1.126 brouard 3910:
1.162 brouard 3911: #ifdef NLOPT
3912: #ifdef NEWUOA
3913: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3914: #else
3915: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3916: #endif
3917: lb=vector(0,npar-1);
3918: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3919: nlopt_set_lower_bounds(opt, lb);
3920: nlopt_set_initial_step1(opt, 0.1);
3921:
3922: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3923: d->function = func;
3924: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3925: nlopt_set_min_objective(opt, myfunc, d);
3926: nlopt_set_xtol_rel(opt, ftol);
3927: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3928: printf("nlopt failed! %d\n",creturn);
3929: }
3930: else {
3931: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3932: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3933: iter=1; /* not equal */
3934: }
3935: nlopt_destroy(opt);
3936: #endif
1.126 brouard 3937: free_matrix(xi,1,npar,1,npar);
3938: fclose(ficrespow);
1.203 brouard 3939: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3940: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3941: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3942:
3943: }
3944:
3945: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3946: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3947: {
3948: double **a,**y,*x,pd;
1.203 brouard 3949: /* double **hess; */
1.164 brouard 3950: int i, j;
1.126 brouard 3951: int *indx;
3952:
3953: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3954: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3955: void lubksb(double **a, int npar, int *indx, double b[]) ;
3956: void ludcmp(double **a, int npar, int *indx, double *d) ;
3957: double gompertz(double p[]);
1.203 brouard 3958: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3959:
3960: printf("\nCalculation of the hessian matrix. Wait...\n");
3961: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3962: for (i=1;i<=npar;i++){
1.203 brouard 3963: printf("%d-",i);fflush(stdout);
3964: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3965:
3966: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3967:
3968: /* printf(" %f ",p[i]);
3969: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3970: }
3971:
3972: for (i=1;i<=npar;i++) {
3973: for (j=1;j<=npar;j++) {
3974: if (j>i) {
1.203 brouard 3975: printf(".%d-%d",i,j);fflush(stdout);
3976: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3977: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3978:
3979: hess[j][i]=hess[i][j];
3980: /*printf(" %lf ",hess[i][j]);*/
3981: }
3982: }
3983: }
3984: printf("\n");
3985: fprintf(ficlog,"\n");
3986:
3987: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3988: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3989:
3990: a=matrix(1,npar,1,npar);
3991: y=matrix(1,npar,1,npar);
3992: x=vector(1,npar);
3993: indx=ivector(1,npar);
3994: for (i=1;i<=npar;i++)
3995: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3996: ludcmp(a,npar,indx,&pd);
3997:
3998: for (j=1;j<=npar;j++) {
3999: for (i=1;i<=npar;i++) x[i]=0;
4000: x[j]=1;
4001: lubksb(a,npar,indx,x);
4002: for (i=1;i<=npar;i++){
4003: matcov[i][j]=x[i];
4004: }
4005: }
4006:
4007: printf("\n#Hessian matrix#\n");
4008: fprintf(ficlog,"\n#Hessian matrix#\n");
4009: for (i=1;i<=npar;i++) {
4010: for (j=1;j<=npar;j++) {
1.203 brouard 4011: printf("%.6e ",hess[i][j]);
4012: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4013: }
4014: printf("\n");
4015: fprintf(ficlog,"\n");
4016: }
4017:
1.203 brouard 4018: /* printf("\n#Covariance matrix#\n"); */
4019: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4020: /* for (i=1;i<=npar;i++) { */
4021: /* for (j=1;j<=npar;j++) { */
4022: /* printf("%.6e ",matcov[i][j]); */
4023: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4024: /* } */
4025: /* printf("\n"); */
4026: /* fprintf(ficlog,"\n"); */
4027: /* } */
4028:
1.126 brouard 4029: /* Recompute Inverse */
1.203 brouard 4030: /* for (i=1;i<=npar;i++) */
4031: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4032: /* ludcmp(a,npar,indx,&pd); */
4033:
4034: /* printf("\n#Hessian matrix recomputed#\n"); */
4035:
4036: /* for (j=1;j<=npar;j++) { */
4037: /* for (i=1;i<=npar;i++) x[i]=0; */
4038: /* x[j]=1; */
4039: /* lubksb(a,npar,indx,x); */
4040: /* for (i=1;i<=npar;i++){ */
4041: /* y[i][j]=x[i]; */
4042: /* printf("%.3e ",y[i][j]); */
4043: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4044: /* } */
4045: /* printf("\n"); */
4046: /* fprintf(ficlog,"\n"); */
4047: /* } */
4048:
4049: /* Verifying the inverse matrix */
4050: #ifdef DEBUGHESS
4051: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4052:
1.203 brouard 4053: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4054: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4055:
4056: for (j=1;j<=npar;j++) {
4057: for (i=1;i<=npar;i++){
1.203 brouard 4058: printf("%.2f ",y[i][j]);
4059: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4060: }
4061: printf("\n");
4062: fprintf(ficlog,"\n");
4063: }
1.203 brouard 4064: #endif
1.126 brouard 4065:
4066: free_matrix(a,1,npar,1,npar);
4067: free_matrix(y,1,npar,1,npar);
4068: free_vector(x,1,npar);
4069: free_ivector(indx,1,npar);
1.203 brouard 4070: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4071:
4072:
4073: }
4074:
4075: /*************** hessian matrix ****************/
4076: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4077: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4078: int i;
4079: int l=1, lmax=20;
1.203 brouard 4080: double k1,k2, res, fx;
1.132 brouard 4081: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4082: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4083: int k=0,kmax=10;
4084: double l1;
4085:
4086: fx=func(x);
4087: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4088: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4089: l1=pow(10,l);
4090: delts=delt;
4091: for(k=1 ; k <kmax; k=k+1){
4092: delt = delta*(l1*k);
4093: p2[theta]=x[theta] +delt;
1.145 brouard 4094: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4095: p2[theta]=x[theta]-delt;
4096: k2=func(p2)-fx;
4097: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4098: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4099:
1.203 brouard 4100: #ifdef DEBUGHESSII
1.126 brouard 4101: 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);
4102: 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);
4103: #endif
4104: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4105: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4106: k=kmax;
4107: }
4108: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4109: k=kmax; l=lmax*10;
1.126 brouard 4110: }
4111: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4112: delts=delt;
4113: }
1.203 brouard 4114: } /* End loop k */
1.126 brouard 4115: }
4116: delti[theta]=delts;
4117: return res;
4118:
4119: }
4120:
1.203 brouard 4121: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4122: {
4123: int i;
1.164 brouard 4124: int l=1, lmax=20;
1.126 brouard 4125: double k1,k2,k3,k4,res,fx;
1.132 brouard 4126: double p2[MAXPARM+1];
1.203 brouard 4127: int k, kmax=1;
4128: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4129:
4130: int firstime=0;
1.203 brouard 4131:
1.126 brouard 4132: fx=func(x);
1.203 brouard 4133: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4134: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4135: p2[thetai]=x[thetai]+delti[thetai]*k;
4136: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4137: k1=func(p2)-fx;
4138:
1.203 brouard 4139: p2[thetai]=x[thetai]+delti[thetai]*k;
4140: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4141: k2=func(p2)-fx;
4142:
1.203 brouard 4143: p2[thetai]=x[thetai]-delti[thetai]*k;
4144: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4145: k3=func(p2)-fx;
4146:
1.203 brouard 4147: p2[thetai]=x[thetai]-delti[thetai]*k;
4148: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4149: k4=func(p2)-fx;
1.203 brouard 4150: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4151: if(k1*k2*k3*k4 <0.){
1.208 brouard 4152: firstime=1;
1.203 brouard 4153: kmax=kmax+10;
1.208 brouard 4154: }
4155: if(kmax >=10 || firstime ==1){
1.246 brouard 4156: 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);
4157: 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 4158: 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);
4159: 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);
4160: }
4161: #ifdef DEBUGHESSIJ
4162: v1=hess[thetai][thetai];
4163: v2=hess[thetaj][thetaj];
4164: cv12=res;
4165: /* Computing eigen value of Hessian matrix */
4166: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4167: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4168: if ((lc2 <0) || (lc1 <0) ){
4169: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4170: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4171: 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);
4172: 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);
4173: }
1.126 brouard 4174: #endif
4175: }
4176: return res;
4177: }
4178:
1.203 brouard 4179: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4180: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4181: /* { */
4182: /* int i; */
4183: /* int l=1, lmax=20; */
4184: /* double k1,k2,k3,k4,res,fx; */
4185: /* double p2[MAXPARM+1]; */
4186: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4187: /* int k=0,kmax=10; */
4188: /* double l1; */
4189:
4190: /* fx=func(x); */
4191: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4192: /* l1=pow(10,l); */
4193: /* delts=delt; */
4194: /* for(k=1 ; k <kmax; k=k+1){ */
4195: /* delt = delti*(l1*k); */
4196: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4197: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4198: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4199: /* k1=func(p2)-fx; */
4200:
4201: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4202: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4203: /* k2=func(p2)-fx; */
4204:
4205: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4206: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4207: /* k3=func(p2)-fx; */
4208:
4209: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4210: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4211: /* k4=func(p2)-fx; */
4212: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4213: /* #ifdef DEBUGHESSIJ */
4214: /* 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); */
4215: /* 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); */
4216: /* #endif */
4217: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4218: /* k=kmax; */
4219: /* } */
4220: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4221: /* k=kmax; l=lmax*10; */
4222: /* } */
4223: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4224: /* delts=delt; */
4225: /* } */
4226: /* } /\* End loop k *\/ */
4227: /* } */
4228: /* delti[theta]=delts; */
4229: /* return res; */
4230: /* } */
4231:
4232:
1.126 brouard 4233: /************** Inverse of matrix **************/
4234: void ludcmp(double **a, int n, int *indx, double *d)
4235: {
4236: int i,imax,j,k;
4237: double big,dum,sum,temp;
4238: double *vv;
4239:
4240: vv=vector(1,n);
4241: *d=1.0;
4242: for (i=1;i<=n;i++) {
4243: big=0.0;
4244: for (j=1;j<=n;j++)
4245: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4246: if (big == 0.0){
4247: printf(" Singular Hessian matrix at row %d:\n",i);
4248: for (j=1;j<=n;j++) {
4249: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4250: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4251: }
4252: fflush(ficlog);
4253: fclose(ficlog);
4254: nrerror("Singular matrix in routine ludcmp");
4255: }
1.126 brouard 4256: vv[i]=1.0/big;
4257: }
4258: for (j=1;j<=n;j++) {
4259: for (i=1;i<j;i++) {
4260: sum=a[i][j];
4261: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4262: a[i][j]=sum;
4263: }
4264: big=0.0;
4265: for (i=j;i<=n;i++) {
4266: sum=a[i][j];
4267: for (k=1;k<j;k++)
4268: sum -= a[i][k]*a[k][j];
4269: a[i][j]=sum;
4270: if ( (dum=vv[i]*fabs(sum)) >= big) {
4271: big=dum;
4272: imax=i;
4273: }
4274: }
4275: if (j != imax) {
4276: for (k=1;k<=n;k++) {
4277: dum=a[imax][k];
4278: a[imax][k]=a[j][k];
4279: a[j][k]=dum;
4280: }
4281: *d = -(*d);
4282: vv[imax]=vv[j];
4283: }
4284: indx[j]=imax;
4285: if (a[j][j] == 0.0) a[j][j]=TINY;
4286: if (j != n) {
4287: dum=1.0/(a[j][j]);
4288: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4289: }
4290: }
4291: free_vector(vv,1,n); /* Doesn't work */
4292: ;
4293: }
4294:
4295: void lubksb(double **a, int n, int *indx, double b[])
4296: {
4297: int i,ii=0,ip,j;
4298: double sum;
4299:
4300: for (i=1;i<=n;i++) {
4301: ip=indx[i];
4302: sum=b[ip];
4303: b[ip]=b[i];
4304: if (ii)
4305: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4306: else if (sum) ii=i;
4307: b[i]=sum;
4308: }
4309: for (i=n;i>=1;i--) {
4310: sum=b[i];
4311: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4312: b[i]=sum/a[i][i];
4313: }
4314: }
4315:
4316: void pstamp(FILE *fichier)
4317: {
1.196 brouard 4318: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4319: }
4320:
1.253 brouard 4321:
4322:
1.126 brouard 4323: /************ Frequencies ********************/
1.251 brouard 4324: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4325: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4326: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4327: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4328:
1.265 brouard 4329: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4330: int iind=0, iage=0;
4331: int mi; /* Effective wave */
4332: int first;
4333: double ***freq; /* Frequencies */
1.268 brouard 4334: double *x, *y, a=0.,b=0.,r=1., sa=0., sb=0.; /* for regression, y=b+m*x and r is the correlation coefficient */
4335: int no=0, linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb);
1.226 brouard 4336: double *meanq;
4337: double **meanqt;
4338: double *pp, **prop, *posprop, *pospropt;
4339: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4340: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4341: double agebegin, ageend;
4342:
4343: pp=vector(1,nlstate);
1.251 brouard 4344: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4345: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4346: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4347: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4348: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4349: meanqt=matrix(1,lastpass,1,nqtveff);
4350: strcpy(fileresp,"P_");
4351: strcat(fileresp,fileresu);
4352: /*strcat(fileresphtm,fileresu);*/
4353: if((ficresp=fopen(fileresp,"w"))==NULL) {
4354: printf("Problem with prevalence resultfile: %s\n", fileresp);
4355: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4356: exit(0);
4357: }
1.240 brouard 4358:
1.226 brouard 4359: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4360: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4361: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4362: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4363: fflush(ficlog);
4364: exit(70);
4365: }
4366: else{
4367: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4368: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4369: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4370: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4371: }
1.237 brouard 4372: 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 4373:
1.226 brouard 4374: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4375: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4376: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4377: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4378: fflush(ficlog);
4379: exit(70);
1.240 brouard 4380: } else{
1.226 brouard 4381: 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 4382: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4383: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4384: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4385: }
1.240 brouard 4386: 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);
4387:
1.253 brouard 4388: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4389: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4390: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4391: j1=0;
1.126 brouard 4392:
1.227 brouard 4393: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4394: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4395: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4396:
4397:
1.226 brouard 4398: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4399: reference=low_education V1=0,V2=0
4400: med_educ V1=1 V2=0,
4401: high_educ V1=0 V2=1
4402: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4403: */
1.249 brouard 4404: dateintsum=0;
4405: k2cpt=0;
4406:
1.253 brouard 4407: if(cptcoveff == 0 )
1.265 brouard 4408: nl=1; /* Constant and age model only */
1.253 brouard 4409: else
4410: nl=2;
1.265 brouard 4411:
4412: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4413: /* Loop on nj=1 or 2 if dummy covariates j!=0
4414: * Loop on j1(1 to 2**cptcoveff) covariate combination
4415: * freq[s1][s2][iage] =0.
4416: * Loop on iind
4417: * ++freq[s1][s2][iage] weighted
4418: * end iind
4419: * if covariate and j!0
4420: * headers Variable on one line
4421: * endif cov j!=0
4422: * header of frequency table by age
4423: * Loop on age
4424: * pp[s1]+=freq[s1][s2][iage] weighted
4425: * pos+=freq[s1][s2][iage] weighted
4426: * Loop on s1 initial state
4427: * fprintf(ficresp
4428: * end s1
4429: * end age
4430: * if j!=0 computes starting values
4431: * end compute starting values
4432: * end j1
4433: * end nl
4434: */
1.253 brouard 4435: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4436: if(nj==1)
4437: j=0; /* First pass for the constant */
1.265 brouard 4438: else{
1.253 brouard 4439: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4440: }
1.251 brouard 4441: first=1;
1.265 brouard 4442: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all covariates combination of the model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 4443: posproptt=0.;
4444: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4445: scanf("%d", i);*/
4446: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4447: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4448: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4449: freq[i][s2][m]=0;
1.251 brouard 4450:
4451: for (i=1; i<=nlstate; i++) {
1.240 brouard 4452: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4453: prop[i][m]=0;
4454: posprop[i]=0;
4455: pospropt[i]=0;
4456: }
4457: /* for (z1=1; z1<= nqfveff; z1++) { */
4458: /* meanq[z1]+=0.; */
4459: /* for(m=1;m<=lastpass;m++){ */
4460: /* meanqt[m][z1]=0.; */
4461: /* } */
4462: /* } */
4463:
4464: /* dateintsum=0; */
4465: /* k2cpt=0; */
4466:
1.265 brouard 4467: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4468: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4469: bool=1;
4470: if(j !=0){
4471: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4472: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4473: /* for (z1=1; z1<= nqfveff; z1++) { */
4474: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4475: /* } */
4476: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4477: /* if(Tvaraff[z1] ==-20){ */
4478: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4479: /* }else if(Tvaraff[z1] ==-10){ */
4480: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4481: /* }else */
4482: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4483: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4484: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4485: /* 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",
4486: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4487: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4488: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4489: } /* Onlyf fixed */
4490: } /* end z1 */
4491: } /* cptcovn > 0 */
4492: } /* end any */
4493: }/* end j==0 */
1.265 brouard 4494: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4495: /* for(m=firstpass; m<=lastpass; m++){ */
4496: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4497: m=mw[mi][iind];
4498: if(j!=0){
4499: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4500: for (z1=1; z1<=cptcoveff; z1++) {
4501: if( Fixed[Tmodelind[z1]]==1){
4502: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4503: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4504: value is -1, we don't select. It differs from the
4505: constant and age model which counts them. */
4506: bool=0; /* not selected */
4507: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4508: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4509: bool=0;
4510: }
4511: }
4512: }
4513: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4514: } /* end j==0 */
4515: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4516: if(bool==1){
4517: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4518: and mw[mi+1][iind]. dh depends on stepm. */
4519: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4520: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4521: if(m >=firstpass && m <=lastpass){
4522: k2=anint[m][iind]+(mint[m][iind]/12.);
4523: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4524: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4525: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4526: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4527: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4528: if (m<lastpass) {
4529: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4530: /* 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]); */
4531: if(s[m][iind]==-1)
4532: 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.));
4533: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4534: /* if((int)agev[m][iind] == 55) */
4535: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4536: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4537: 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 4538: }
1.251 brouard 4539: } /* end if between passes */
4540: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4541: dateintsum=dateintsum+k2; /* on all covariates ?*/
4542: k2cpt++;
4543: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4544: }
1.251 brouard 4545: }else{
4546: bool=1;
4547: }/* end bool 2 */
4548: } /* end m */
4549: } /* end bool */
4550: } /* end iind = 1 to imx */
4551: /* prop[s][age] is feeded for any initial and valid live state as well as
4552: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4553:
4554:
4555: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4556: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4557: pstamp(ficresp);
1.251 brouard 4558: if (cptcoveff>0 && j!=0){
1.265 brouard 4559: pstamp(ficresp);
1.251 brouard 4560: printf( "\n#********** Variable ");
4561: fprintf(ficresp, "\n#********** Variable ");
4562: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4563: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4564: fprintf(ficlog, "\n#********** Variable ");
4565: for (z1=1; z1<=cptcoveff; z1++){
4566: if(!FixedV[Tvaraff[z1]]){
4567: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4568: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4569: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4570: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4571: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4572: }else{
1.251 brouard 4573: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4574: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4575: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4576: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4577: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4578: }
4579: }
4580: printf( "**********\n#");
4581: fprintf(ficresp, "**********\n#");
4582: fprintf(ficresphtm, "**********</h3>\n");
4583: fprintf(ficresphtmfr, "**********</h3>\n");
4584: fprintf(ficlog, "**********\n");
4585: }
4586: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4587: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4588: fprintf(ficresp, " Age");
4589: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 4590: for(i=1; i<=nlstate;i++) {
1.265 brouard 4591: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4592: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4593: }
1.265 brouard 4594: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4595: fprintf(ficresphtm, "\n");
4596:
4597: /* Header of frequency table by age */
4598: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4599: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4600: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4601: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4602: if(s2!=0 && m!=0)
4603: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4604: }
1.226 brouard 4605: }
1.251 brouard 4606: fprintf(ficresphtmfr, "\n");
4607:
4608: /* For each age */
4609: for(iage=iagemin; iage <= iagemax+3; iage++){
4610: fprintf(ficresphtm,"<tr>");
4611: if(iage==iagemax+1){
4612: fprintf(ficlog,"1");
4613: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4614: }else if(iage==iagemax+2){
4615: fprintf(ficlog,"0");
4616: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4617: }else if(iage==iagemax+3){
4618: fprintf(ficlog,"Total");
4619: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4620: }else{
1.240 brouard 4621: if(first==1){
1.251 brouard 4622: first=0;
4623: printf("See log file for details...\n");
4624: }
4625: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4626: fprintf(ficlog,"Age %d", iage);
4627: }
1.265 brouard 4628: for(s1=1; s1 <=nlstate ; s1++){
4629: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4630: pp[s1] += freq[s1][m][iage];
1.251 brouard 4631: }
1.265 brouard 4632: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4633: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4634: pos += freq[s1][m][iage];
4635: if(pp[s1]>=1.e-10){
1.251 brouard 4636: if(first==1){
1.265 brouard 4637: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4638: }
1.265 brouard 4639: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4640: }else{
4641: if(first==1)
1.265 brouard 4642: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4643: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4644: }
4645: }
4646:
1.265 brouard 4647: for(s1=1; s1 <=nlstate ; s1++){
4648: /* posprop[s1]=0; */
4649: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4650: pp[s1] += freq[s1][m][iage];
4651: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4652:
4653: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4654: pos += pp[s1]; /* pos is the total number of transitions until this age */
4655: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4656: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4657: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4658: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4659: }
4660:
4661: /* Writing ficresp */
4662: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4663: if( iage <= iagemax){
4664: fprintf(ficresp," %d",iage);
4665: }
4666: }else if( nj==2){
4667: if( iage <= iagemax){
4668: fprintf(ficresp," %d",iage);
4669: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4670: }
1.240 brouard 4671: }
1.265 brouard 4672: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4673: if(pos>=1.e-5){
1.251 brouard 4674: if(first==1)
1.265 brouard 4675: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4676: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4677: }else{
4678: if(first==1)
1.265 brouard 4679: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4680: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4681: }
4682: if( iage <= iagemax){
4683: if(pos>=1.e-5){
1.265 brouard 4684: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4685: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4686: }else if( nj==2){
4687: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4688: }
4689: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4690: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4691: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4692: } else{
4693: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4694: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4695: }
1.240 brouard 4696: }
1.265 brouard 4697: pospropt[s1] +=posprop[s1];
4698: } /* end loop s1 */
1.251 brouard 4699: /* pospropt=0.; */
1.265 brouard 4700: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4701: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4702: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4703: if(first==1){
1.265 brouard 4704: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4705: }
1.265 brouard 4706: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4707: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4708: }
1.265 brouard 4709: if(s1!=0 && m!=0)
4710: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4711: }
1.265 brouard 4712: } /* end loop s1 */
1.251 brouard 4713: posproptt=0.;
1.265 brouard 4714: for(s1=1; s1 <=nlstate; s1++){
4715: posproptt += pospropt[s1];
1.251 brouard 4716: }
4717: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4718: fprintf(ficresphtm,"</tr>\n");
4719: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4720: if(iage <= iagemax)
4721: fprintf(ficresp,"\n");
1.240 brouard 4722: }
1.251 brouard 4723: if(first==1)
4724: printf("Others in log...\n");
4725: fprintf(ficlog,"\n");
4726: } /* end loop age iage */
1.265 brouard 4727:
1.251 brouard 4728: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4729: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4730: if(posproptt < 1.e-5){
1.265 brouard 4731: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4732: }else{
1.265 brouard 4733: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4734: }
1.226 brouard 4735: }
1.251 brouard 4736: fprintf(ficresphtm,"</tr>\n");
4737: fprintf(ficresphtm,"</table>\n");
4738: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4739: if(posproptt < 1.e-5){
1.251 brouard 4740: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4741: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4742: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4743: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4744: invalidvarcomb[j1]=1;
1.226 brouard 4745: }else{
1.251 brouard 4746: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4747: invalidvarcomb[j1]=0;
1.226 brouard 4748: }
1.251 brouard 4749: fprintf(ficresphtmfr,"</table>\n");
4750: fprintf(ficlog,"\n");
4751: if(j!=0){
4752: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4753: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4754: for(k=1; k <=(nlstate+ndeath); k++){
4755: if (k != i) {
1.265 brouard 4756: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4757: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4758: if(j1==1){ /* All dummy covariates to zero */
4759: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4760: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4761: printf("%d%d ",i,k);
4762: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4763: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
4764: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4765: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4766: }
1.253 brouard 4767: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4768: for(iage=iagemin; iage <= iagemax+3; iage++){
4769: x[iage]= (double)iage;
4770: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4771: /* printf("i=%d, k=%d, s1=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,s1,j1,jj, iage, y[iage]); */
1.253 brouard 4772: }
1.268 brouard 4773: /* Some are not finite, but linreg will ignore these ages */
4774: no=0;
1.253 brouard 4775: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4776: pstart[s1]=b;
4777: pstart[s1-1]=a;
1.252 brouard 4778: }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 */
4779: 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]);
4780: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.265 brouard 4781: pstart[s1]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4782: printf("%d%d ",i,k);
4783: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4784: printf("s1=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",s1,i,k,s1,p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
1.251 brouard 4785: }else{ /* Other cases, like quantitative fixed or varying covariates */
4786: ;
4787: }
4788: /* printf("%12.7f )", param[i][jj][k]); */
4789: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4790: s1++;
1.251 brouard 4791: } /* end jj */
4792: } /* end k!= i */
4793: } /* end k */
1.265 brouard 4794: } /* end i, s1 */
1.251 brouard 4795: } /* end j !=0 */
4796: } /* end selected combination of covariate j1 */
4797: if(j==0){ /* We can estimate starting values from the occurences in each case */
4798: printf("#Freqsummary: Starting values for the constants:\n");
4799: fprintf(ficlog,"\n");
1.265 brouard 4800: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4801: for(k=1; k <=(nlstate+ndeath); k++){
4802: if (k != i) {
4803: printf("%d%d ",i,k);
4804: fprintf(ficlog,"%d%d ",i,k);
4805: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4806: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4807: if(jj==1){ /* Age has to be done */
1.265 brouard 4808: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4809: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4810: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
1.251 brouard 4811: }
4812: /* printf("%12.7f )", param[i][jj][k]); */
4813: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4814: s1++;
1.250 brouard 4815: }
1.251 brouard 4816: printf("\n");
4817: fprintf(ficlog,"\n");
1.250 brouard 4818: }
4819: }
4820: }
1.251 brouard 4821: printf("#Freqsummary\n");
4822: fprintf(ficlog,"\n");
1.265 brouard 4823: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4824: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4825: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4826: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4827: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4828: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4829: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4830: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4831: /* } */
4832: }
1.265 brouard 4833: } /* end loop s1 */
1.251 brouard 4834:
4835: printf("\n");
4836: fprintf(ficlog,"\n");
4837: } /* end j=0 */
1.249 brouard 4838: } /* end j */
1.252 brouard 4839:
1.253 brouard 4840: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4841: for(i=1, jk=1; i <=nlstate; i++){
4842: for(j=1; j <=nlstate+ndeath; j++){
4843: if(j!=i){
4844: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4845: printf("%1d%1d",i,j);
4846: fprintf(ficparo,"%1d%1d",i,j);
4847: for(k=1; k<=ncovmodel;k++){
4848: /* printf(" %lf",param[i][j][k]); */
4849: /* fprintf(ficparo," %lf",param[i][j][k]); */
4850: p[jk]=pstart[jk];
4851: printf(" %f ",pstart[jk]);
4852: fprintf(ficparo," %f ",pstart[jk]);
4853: jk++;
4854: }
4855: printf("\n");
4856: fprintf(ficparo,"\n");
4857: }
4858: }
4859: }
4860: } /* end mle=-2 */
1.226 brouard 4861: dateintmean=dateintsum/k2cpt;
1.240 brouard 4862:
1.226 brouard 4863: fclose(ficresp);
4864: fclose(ficresphtm);
4865: fclose(ficresphtmfr);
4866: free_vector(meanq,1,nqfveff);
4867: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4868: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4869: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4870: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4871: free_vector(pospropt,1,nlstate);
4872: free_vector(posprop,1,nlstate);
1.251 brouard 4873: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4874: free_vector(pp,1,nlstate);
4875: /* End of freqsummary */
4876: }
1.126 brouard 4877:
1.268 brouard 4878: /* Simple linear regression */
4879: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4880:
4881: /* y=a+bx regression */
4882: double sumx = 0.0; /* sum of x */
4883: double sumx2 = 0.0; /* sum of x**2 */
4884: double sumxy = 0.0; /* sum of x * y */
4885: double sumy = 0.0; /* sum of y */
4886: double sumy2 = 0.0; /* sum of y**2 */
4887: double sume2 = 0.0; /* sum of square or residuals */
4888: double yhat;
4889:
4890: double denom=0;
4891: int i;
4892: int ne=*no;
4893:
4894: for ( i=ifi, ne=0;i<=ila;i++) {
4895: if(!isfinite(x[i]) || !isfinite(y[i])){
4896: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4897: continue;
4898: }
4899: ne=ne+1;
4900: sumx += x[i];
4901: sumx2 += x[i]*x[i];
4902: sumxy += x[i] * y[i];
4903: sumy += y[i];
4904: sumy2 += y[i]*y[i];
4905: denom = (ne * sumx2 - sumx*sumx);
4906: /* 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); */
4907: }
4908:
4909: denom = (ne * sumx2 - sumx*sumx);
4910: if (denom == 0) {
4911: // vertical, slope m is infinity
4912: *b = INFINITY;
4913: *a = 0;
4914: if (r) *r = 0;
4915: return 1;
4916: }
4917:
4918: *b = (ne * sumxy - sumx * sumy) / denom;
4919: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4920: if (r!=NULL) {
4921: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4922: sqrt((sumx2 - sumx*sumx/ne) *
4923: (sumy2 - sumy*sumy/ne));
4924: }
4925: *no=ne;
4926: for ( i=ifi, ne=0;i<=ila;i++) {
4927: if(!isfinite(x[i]) || !isfinite(y[i])){
4928: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4929: continue;
4930: }
4931: ne=ne+1;
4932: yhat = y[i] - *a -*b* x[i];
4933: sume2 += yhat * yhat ;
4934:
4935: denom = (ne * sumx2 - sumx*sumx);
4936: /* 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); */
4937: }
4938: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4939: *sa= *sb * sqrt(sumx2/ne);
4940:
4941: return 0;
4942: }
4943:
1.126 brouard 4944: /************ Prevalence ********************/
1.227 brouard 4945: 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)
4946: {
4947: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4948: in each health status at the date of interview (if between dateprev1 and dateprev2).
4949: We still use firstpass and lastpass as another selection.
4950: */
1.126 brouard 4951:
1.227 brouard 4952: int i, m, jk, j1, bool, z1,j, iv;
4953: int mi; /* Effective wave */
4954: int iage;
4955: double agebegin, ageend;
4956:
4957: double **prop;
4958: double posprop;
4959: double y2; /* in fractional years */
4960: int iagemin, iagemax;
4961: int first; /** to stop verbosity which is redirected to log file */
4962:
4963: iagemin= (int) agemin;
4964: iagemax= (int) agemax;
4965: /*pp=vector(1,nlstate);*/
1.251 brouard 4966: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4967: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4968: j1=0;
1.222 brouard 4969:
1.227 brouard 4970: /*j=cptcoveff;*/
4971: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4972:
1.227 brouard 4973: first=1;
4974: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4975: for (i=1; i<=nlstate; i++)
1.251 brouard 4976: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4977: prop[i][iage]=0.0;
4978: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4979: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4980: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4981:
4982: for (i=1; i<=imx; i++) { /* Each individual */
4983: bool=1;
4984: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4985: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4986: m=mw[mi][i];
4987: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4988: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4989: for (z1=1; z1<=cptcoveff; z1++){
4990: if( Fixed[Tmodelind[z1]]==1){
4991: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4992: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4993: bool=0;
4994: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4995: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4996: bool=0;
4997: }
4998: }
4999: if(bool==1){ /* Otherwise we skip that wave/person */
5000: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5001: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5002: if(m >=firstpass && m <=lastpass){
5003: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5004: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5005: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5006: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5007: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5008: 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);
5009: exit(1);
5010: }
5011: if (s[m][i]>0 && s[m][i]<=nlstate) {
5012: /*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]]);*/
5013: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5014: prop[s[m][i]][iagemax+3] += weight[i];
5015: } /* end valid statuses */
5016: } /* end selection of dates */
5017: } /* end selection of waves */
5018: } /* end bool */
5019: } /* end wave */
5020: } /* end individual */
5021: for(i=iagemin; i <= iagemax+3; i++){
5022: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5023: posprop += prop[jk][i];
5024: }
5025:
5026: for(jk=1; jk <=nlstate ; jk++){
5027: if( i <= iagemax){
5028: if(posprop>=1.e-5){
5029: probs[i][jk][j1]= prop[jk][i]/posprop;
5030: } else{
5031: if(first==1){
5032: first=0;
1.266 brouard 5033: printf("Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
5034: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
5035: }else{
5036: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 5037: }
5038: }
5039: }
5040: }/* end jk */
5041: }/* end i */
1.222 brouard 5042: /*} *//* end i1 */
1.227 brouard 5043: } /* end j1 */
1.222 brouard 5044:
1.227 brouard 5045: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5046: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5047: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5048: } /* End of prevalence */
1.126 brouard 5049:
5050: /************* Waves Concatenation ***************/
5051:
5052: 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)
5053: {
5054: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5055: Death is a valid wave (if date is known).
5056: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5057: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5058: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5059: */
1.126 brouard 5060:
1.224 brouard 5061: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5062: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5063: double sum=0., jmean=0.;*/
1.224 brouard 5064: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5065: int j, k=0,jk, ju, jl;
5066: double sum=0.;
5067: first=0;
1.214 brouard 5068: firstwo=0;
1.217 brouard 5069: firsthree=0;
1.218 brouard 5070: firstfour=0;
1.164 brouard 5071: jmin=100000;
1.126 brouard 5072: jmax=-1;
5073: jmean=0.;
1.224 brouard 5074:
5075: /* Treating live states */
1.214 brouard 5076: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5077: mi=0; /* First valid wave */
1.227 brouard 5078: mli=0; /* Last valid wave */
1.126 brouard 5079: m=firstpass;
1.214 brouard 5080: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5081: 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 */
5082: mli=m-1;/* mw[++mi][i]=m-1; */
5083: }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 */
5084: mw[++mi][i]=m;
5085: mli=m;
1.224 brouard 5086: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5087: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5088: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5089: }
1.227 brouard 5090: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5091: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5092: break;
1.224 brouard 5093: #else
1.227 brouard 5094: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5095: if(firsthree == 0){
1.262 brouard 5096: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p%d%d .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 5097: firsthree=1;
5098: }
1.262 brouard 5099: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p%d%d .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 5100: mw[++mi][i]=m;
5101: mli=m;
5102: }
5103: if(s[m][i]==-2){ /* Vital status is really unknown */
5104: nbwarn++;
5105: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5106: 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);
5107: 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);
5108: }
5109: break;
5110: }
5111: break;
1.224 brouard 5112: #endif
1.227 brouard 5113: }/* End m >= lastpass */
1.126 brouard 5114: }/* end while */
1.224 brouard 5115:
1.227 brouard 5116: /* 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 5117: /* After last pass */
1.224 brouard 5118: /* Treating death states */
1.214 brouard 5119: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5120: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5121: /* } */
1.126 brouard 5122: mi++; /* Death is another wave */
5123: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5124: /* Only death is a correct wave */
1.126 brouard 5125: mw[mi][i]=m;
1.257 brouard 5126: } /* else not in a death state */
1.224 brouard 5127: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5128: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5129: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5130: 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 */
5131: nbwarn++;
5132: if(firstfiv==0){
5133: 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 );
5134: firstfiv=1;
5135: }else{
5136: 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 );
5137: }
5138: }else{ /* Death occured afer last wave potential bias */
5139: nberr++;
5140: if(firstwo==0){
1.257 brouard 5141: 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 5142: firstwo=1;
5143: }
1.257 brouard 5144: 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 5145: }
1.257 brouard 5146: }else{ /* if date of interview is unknown */
1.227 brouard 5147: /* death is known but not confirmed by death status at any wave */
5148: if(firstfour==0){
5149: 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 );
5150: firstfour=1;
5151: }
5152: 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 5153: }
1.224 brouard 5154: } /* end if date of death is known */
5155: #endif
5156: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5157: /* wav[i]=mw[mi][i]; */
1.126 brouard 5158: if(mi==0){
5159: nbwarn++;
5160: if(first==0){
1.227 brouard 5161: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5162: first=1;
1.126 brouard 5163: }
5164: if(first==1){
1.227 brouard 5165: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5166: }
5167: } /* end mi==0 */
5168: } /* End individuals */
1.214 brouard 5169: /* wav and mw are no more changed */
1.223 brouard 5170:
1.214 brouard 5171:
1.126 brouard 5172: for(i=1; i<=imx; i++){
5173: for(mi=1; mi<wav[i];mi++){
5174: if (stepm <=0)
1.227 brouard 5175: dh[mi][i]=1;
1.126 brouard 5176: else{
1.260 brouard 5177: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5178: if (agedc[i] < 2*AGESUP) {
5179: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5180: if(j==0) j=1; /* Survives at least one month after exam */
5181: else if(j<0){
5182: nberr++;
5183: 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]);
5184: j=1; /* Temporary Dangerous patch */
5185: 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);
5186: 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]);
5187: 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);
5188: }
5189: k=k+1;
5190: if (j >= jmax){
5191: jmax=j;
5192: ijmax=i;
5193: }
5194: if (j <= jmin){
5195: jmin=j;
5196: ijmin=i;
5197: }
5198: sum=sum+j;
5199: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5200: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5201: }
5202: }
5203: else{
5204: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5205: /* 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 5206:
1.227 brouard 5207: k=k+1;
5208: if (j >= jmax) {
5209: jmax=j;
5210: ijmax=i;
5211: }
5212: else if (j <= jmin){
5213: jmin=j;
5214: ijmin=i;
5215: }
5216: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5217: /*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]);*/
5218: if(j<0){
5219: nberr++;
5220: 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]);
5221: 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]);
5222: }
5223: sum=sum+j;
5224: }
5225: jk= j/stepm;
5226: jl= j -jk*stepm;
5227: ju= j -(jk+1)*stepm;
5228: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5229: if(jl==0){
5230: dh[mi][i]=jk;
5231: bh[mi][i]=0;
5232: }else{ /* We want a negative bias in order to only have interpolation ie
5233: * to avoid the price of an extra matrix product in likelihood */
5234: dh[mi][i]=jk+1;
5235: bh[mi][i]=ju;
5236: }
5237: }else{
5238: if(jl <= -ju){
5239: dh[mi][i]=jk;
5240: bh[mi][i]=jl; /* bias is positive if real duration
5241: * is higher than the multiple of stepm and negative otherwise.
5242: */
5243: }
5244: else{
5245: dh[mi][i]=jk+1;
5246: bh[mi][i]=ju;
5247: }
5248: if(dh[mi][i]==0){
5249: dh[mi][i]=1; /* At least one step */
5250: bh[mi][i]=ju; /* At least one step */
5251: /* 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);*/
5252: }
5253: } /* end if mle */
1.126 brouard 5254: }
5255: } /* end wave */
5256: }
5257: jmean=sum/k;
5258: 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 5259: 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 5260: }
1.126 brouard 5261:
5262: /*********** Tricode ****************************/
1.220 brouard 5263: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5264: {
5265: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5266: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5267: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5268: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5269: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5270: */
1.130 brouard 5271:
1.242 brouard 5272: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5273: int modmaxcovj=0; /* Modality max of covariates j */
5274: int cptcode=0; /* Modality max of covariates j */
5275: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5276:
5277:
1.242 brouard 5278: /* cptcoveff=0; */
5279: /* *cptcov=0; */
1.126 brouard 5280:
1.242 brouard 5281: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5282:
1.242 brouard 5283: /* Loop on covariates without age and products and no quantitative variable */
5284: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5285: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5286: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5287: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5288: switch(Fixed[k]) {
5289: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5290: 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*/
5291: ij=(int)(covar[Tvar[k]][i]);
5292: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5293: * If product of Vn*Vm, still boolean *:
5294: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5295: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5296: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5297: modality of the nth covariate of individual i. */
5298: if (ij > modmaxcovj)
5299: modmaxcovj=ij;
5300: else if (ij < modmincovj)
5301: modmincovj=ij;
5302: if ((ij < -1) && (ij > NCOVMAX)){
5303: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5304: exit(1);
5305: }else
5306: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5307: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5308: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5309: /* getting the maximum value of the modality of the covariate
5310: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5311: female ies 1, then modmaxcovj=1.
5312: */
5313: } /* end for loop on individuals i */
5314: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5315: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5316: cptcode=modmaxcovj;
5317: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5318: /*for (i=0; i<=cptcode; i++) {*/
5319: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5320: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5321: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5322: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5323: if( j != -1){
5324: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5325: covariate for which somebody answered excluding
5326: undefined. Usually 2: 0 and 1. */
5327: }
5328: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5329: covariate for which somebody answered including
5330: undefined. Usually 3: -1, 0 and 1. */
5331: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5332: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5333: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5334:
1.242 brouard 5335: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5336: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5337: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5338: /* modmincovj=3; modmaxcovj = 7; */
5339: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5340: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5341: /* defining two dummy variables: variables V1_1 and V1_2.*/
5342: /* nbcode[Tvar[j]][ij]=k; */
5343: /* nbcode[Tvar[j]][1]=0; */
5344: /* nbcode[Tvar[j]][2]=1; */
5345: /* nbcode[Tvar[j]][3]=2; */
5346: /* To be continued (not working yet). */
5347: ij=0; /* ij is similar to i but can jump over null modalities */
5348: 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*/
5349: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5350: break;
5351: }
5352: ij++;
5353: 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*/
5354: cptcode = ij; /* New max modality for covar j */
5355: } /* end of loop on modality i=-1 to 1 or more */
5356: break;
5357: case 1: /* Testing on varying covariate, could be simple and
5358: * should look at waves or product of fixed *
5359: * varying. No time to test -1, assuming 0 and 1 only */
5360: ij=0;
5361: for(i=0; i<=1;i++){
5362: nbcode[Tvar[k]][++ij]=i;
5363: }
5364: break;
5365: default:
5366: break;
5367: } /* end switch */
5368: } /* end dummy test */
5369:
5370: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5371: /* /\*recode from 0 *\/ */
5372: /* k is a modality. If we have model=V1+V1*sex */
5373: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5374: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5375: /* } */
5376: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5377: /* if (ij > ncodemax[j]) { */
5378: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5379: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5380: /* break; */
5381: /* } */
5382: /* } /\* end of loop on modality k *\/ */
5383: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5384:
5385: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5386: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5387: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5388: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5389: 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 */
5390: 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 */
5391: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5392: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5393:
5394: ij=0;
5395: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5396: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5397: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5398: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5399: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5400: /* If product not in single variable we don't print results */
5401: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5402: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5403: 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*/
5404: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5405: 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 */
5406: if(Fixed[k]!=0)
5407: anyvaryingduminmodel=1;
5408: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5409: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5410: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5411: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5412: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5413: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5414: }
5415: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5416: /* ij--; */
5417: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5418: *cptcov=ij; /*Number of total real effective covariates: effective
5419: * because they can be excluded from the model and real
5420: * if in the model but excluded because missing values, but how to get k from ij?*/
5421: for(j=ij+1; j<= cptcovt; j++){
5422: Tvaraff[j]=0;
5423: Tmodelind[j]=0;
5424: }
5425: for(j=ntveff+1; j<= cptcovt; j++){
5426: TmodelInvind[j]=0;
5427: }
5428: /* To be sorted */
5429: ;
5430: }
1.126 brouard 5431:
1.145 brouard 5432:
1.126 brouard 5433: /*********** Health Expectancies ****************/
5434:
1.235 brouard 5435: 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 5436:
5437: {
5438: /* Health expectancies, no variances */
1.164 brouard 5439: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5440: int nhstepma, nstepma; /* Decreasing with age */
5441: double age, agelim, hf;
5442: double ***p3mat;
5443: double eip;
5444:
1.238 brouard 5445: /* pstamp(ficreseij); */
1.126 brouard 5446: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5447: fprintf(ficreseij,"# Age");
5448: for(i=1; i<=nlstate;i++){
5449: for(j=1; j<=nlstate;j++){
5450: fprintf(ficreseij," e%1d%1d ",i,j);
5451: }
5452: fprintf(ficreseij," e%1d. ",i);
5453: }
5454: fprintf(ficreseij,"\n");
5455:
5456:
5457: if(estepm < stepm){
5458: printf ("Problem %d lower than %d\n",estepm, stepm);
5459: }
5460: else hstepm=estepm;
5461: /* We compute the life expectancy from trapezoids spaced every estepm months
5462: * This is mainly to measure the difference between two models: for example
5463: * if stepm=24 months pijx are given only every 2 years and by summing them
5464: * we are calculating an estimate of the Life Expectancy assuming a linear
5465: * progression in between and thus overestimating or underestimating according
5466: * to the curvature of the survival function. If, for the same date, we
5467: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5468: * to compare the new estimate of Life expectancy with the same linear
5469: * hypothesis. A more precise result, taking into account a more precise
5470: * curvature will be obtained if estepm is as small as stepm. */
5471:
5472: /* For example we decided to compute the life expectancy with the smallest unit */
5473: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5474: nhstepm is the number of hstepm from age to agelim
5475: nstepm is the number of stepm from age to agelin.
1.270 brouard 5476: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5477: and note for a fixed period like estepm months */
5478: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5479: survival function given by stepm (the optimization length). Unfortunately it
5480: means that if the survival funtion is printed only each two years of age and if
5481: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5482: results. So we changed our mind and took the option of the best precision.
5483: */
5484: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5485:
5486: agelim=AGESUP;
5487: /* If stepm=6 months */
5488: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5489: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5490:
5491: /* nhstepm age range expressed in number of stepm */
5492: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5493: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5494: /* if (stepm >= YEARM) hstepm=1;*/
5495: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5496: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5497:
5498: for (age=bage; age<=fage; age ++){
5499: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5500: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5501: /* if (stepm >= YEARM) hstepm=1;*/
5502: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5503:
5504: /* If stepm=6 months */
5505: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5506: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5507:
1.235 brouard 5508: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5509:
5510: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5511:
5512: printf("%d|",(int)age);fflush(stdout);
5513: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5514:
5515: /* Computing expectancies */
5516: for(i=1; i<=nlstate;i++)
5517: for(j=1; j<=nlstate;j++)
5518: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5519: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5520:
5521: /* 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]);*/
5522:
5523: }
5524:
5525: fprintf(ficreseij,"%3.0f",age );
5526: for(i=1; i<=nlstate;i++){
5527: eip=0;
5528: for(j=1; j<=nlstate;j++){
5529: eip +=eij[i][j][(int)age];
5530: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5531: }
5532: fprintf(ficreseij,"%9.4f", eip );
5533: }
5534: fprintf(ficreseij,"\n");
5535:
5536: }
5537: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5538: printf("\n");
5539: fprintf(ficlog,"\n");
5540:
5541: }
5542:
1.235 brouard 5543: 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 5544:
5545: {
5546: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5547: to initial status i, ei. .
1.126 brouard 5548: */
5549: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5550: int nhstepma, nstepma; /* Decreasing with age */
5551: double age, agelim, hf;
5552: double ***p3matp, ***p3matm, ***varhe;
5553: double **dnewm,**doldm;
5554: double *xp, *xm;
5555: double **gp, **gm;
5556: double ***gradg, ***trgradg;
5557: int theta;
5558:
5559: double eip, vip;
5560:
5561: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5562: xp=vector(1,npar);
5563: xm=vector(1,npar);
5564: dnewm=matrix(1,nlstate*nlstate,1,npar);
5565: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5566:
5567: pstamp(ficresstdeij);
5568: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5569: fprintf(ficresstdeij,"# Age");
5570: for(i=1; i<=nlstate;i++){
5571: for(j=1; j<=nlstate;j++)
5572: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5573: fprintf(ficresstdeij," e%1d. ",i);
5574: }
5575: fprintf(ficresstdeij,"\n");
5576:
5577: pstamp(ficrescveij);
5578: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5579: fprintf(ficrescveij,"# Age");
5580: for(i=1; i<=nlstate;i++)
5581: for(j=1; j<=nlstate;j++){
5582: cptj= (j-1)*nlstate+i;
5583: for(i2=1; i2<=nlstate;i2++)
5584: for(j2=1; j2<=nlstate;j2++){
5585: cptj2= (j2-1)*nlstate+i2;
5586: if(cptj2 <= cptj)
5587: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5588: }
5589: }
5590: fprintf(ficrescveij,"\n");
5591:
5592: if(estepm < stepm){
5593: printf ("Problem %d lower than %d\n",estepm, stepm);
5594: }
5595: else hstepm=estepm;
5596: /* We compute the life expectancy from trapezoids spaced every estepm months
5597: * This is mainly to measure the difference between two models: for example
5598: * if stepm=24 months pijx are given only every 2 years and by summing them
5599: * we are calculating an estimate of the Life Expectancy assuming a linear
5600: * progression in between and thus overestimating or underestimating according
5601: * to the curvature of the survival function. If, for the same date, we
5602: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5603: * to compare the new estimate of Life expectancy with the same linear
5604: * hypothesis. A more precise result, taking into account a more precise
5605: * curvature will be obtained if estepm is as small as stepm. */
5606:
5607: /* For example we decided to compute the life expectancy with the smallest unit */
5608: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5609: nhstepm is the number of hstepm from age to agelim
5610: nstepm is the number of stepm from age to agelin.
5611: Look at hpijx to understand the reason of that which relies in memory size
5612: and note for a fixed period like estepm months */
5613: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5614: survival function given by stepm (the optimization length). Unfortunately it
5615: means that if the survival funtion is printed only each two years of age and if
5616: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5617: results. So we changed our mind and took the option of the best precision.
5618: */
5619: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5620:
5621: /* If stepm=6 months */
5622: /* nhstepm age range expressed in number of stepm */
5623: agelim=AGESUP;
5624: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5625: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5626: /* if (stepm >= YEARM) hstepm=1;*/
5627: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5628:
5629: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5630: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5631: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5632: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5633: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5634: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5635:
5636: for (age=bage; age<=fage; age ++){
5637: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5638: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5639: /* if (stepm >= YEARM) hstepm=1;*/
5640: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5641:
1.126 brouard 5642: /* If stepm=6 months */
5643: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5644: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5645:
5646: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5647:
1.126 brouard 5648: /* Computing Variances of health expectancies */
5649: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5650: decrease memory allocation */
5651: for(theta=1; theta <=npar; theta++){
5652: for(i=1; i<=npar; i++){
1.222 brouard 5653: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5654: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5655: }
1.235 brouard 5656: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5657: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5658:
1.126 brouard 5659: for(j=1; j<= nlstate; j++){
1.222 brouard 5660: for(i=1; i<=nlstate; i++){
5661: for(h=0; h<=nhstepm-1; h++){
5662: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5663: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5664: }
5665: }
1.126 brouard 5666: }
1.218 brouard 5667:
1.126 brouard 5668: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5669: for(h=0; h<=nhstepm-1; h++){
5670: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5671: }
1.126 brouard 5672: }/* End theta */
5673:
5674:
5675: for(h=0; h<=nhstepm-1; h++)
5676: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5677: for(theta=1; theta <=npar; theta++)
5678: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5679:
1.218 brouard 5680:
1.222 brouard 5681: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5682: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5683: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5684:
1.222 brouard 5685: printf("%d|",(int)age);fflush(stdout);
5686: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5687: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5688: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5689: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5690: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5691: for(ij=1;ij<=nlstate*nlstate;ij++)
5692: for(ji=1;ji<=nlstate*nlstate;ji++)
5693: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5694: }
5695: }
1.218 brouard 5696:
1.126 brouard 5697: /* Computing expectancies */
1.235 brouard 5698: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5699: for(i=1; i<=nlstate;i++)
5700: for(j=1; j<=nlstate;j++)
1.222 brouard 5701: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5702: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5703:
1.222 brouard 5704: /* 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 5705:
1.222 brouard 5706: }
1.269 brouard 5707:
5708: /* Standard deviation of expectancies ij */
1.126 brouard 5709: fprintf(ficresstdeij,"%3.0f",age );
5710: for(i=1; i<=nlstate;i++){
5711: eip=0.;
5712: vip=0.;
5713: for(j=1; j<=nlstate;j++){
1.222 brouard 5714: eip += eij[i][j][(int)age];
5715: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5716: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5717: 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 5718: }
5719: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5720: }
5721: fprintf(ficresstdeij,"\n");
1.218 brouard 5722:
1.269 brouard 5723: /* Variance of expectancies ij */
1.126 brouard 5724: fprintf(ficrescveij,"%3.0f",age );
5725: for(i=1; i<=nlstate;i++)
5726: for(j=1; j<=nlstate;j++){
1.222 brouard 5727: cptj= (j-1)*nlstate+i;
5728: for(i2=1; i2<=nlstate;i2++)
5729: for(j2=1; j2<=nlstate;j2++){
5730: cptj2= (j2-1)*nlstate+i2;
5731: if(cptj2 <= cptj)
5732: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5733: }
1.126 brouard 5734: }
5735: fprintf(ficrescveij,"\n");
1.218 brouard 5736:
1.126 brouard 5737: }
5738: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5739: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5740: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5741: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5742: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5743: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5744: printf("\n");
5745: fprintf(ficlog,"\n");
1.218 brouard 5746:
1.126 brouard 5747: free_vector(xm,1,npar);
5748: free_vector(xp,1,npar);
5749: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5750: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5751: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5752: }
1.218 brouard 5753:
1.126 brouard 5754: /************ Variance ******************/
1.235 brouard 5755: 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 5756: {
5757: /* Variance of health expectancies */
5758: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5759: /* double **newm;*/
5760: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5761:
5762: /* int movingaverage(); */
5763: double **dnewm,**doldm;
5764: double **dnewmp,**doldmp;
5765: int i, j, nhstepm, hstepm, h, nstepm ;
5766: int k;
5767: double *xp;
5768: double **gp, **gm; /* for var eij */
5769: double ***gradg, ***trgradg; /*for var eij */
5770: double **gradgp, **trgradgp; /* for var p point j */
5771: double *gpp, *gmp; /* for var p point j */
5772: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5773: double ***p3mat;
5774: double age,agelim, hf;
5775: /* double ***mobaverage; */
5776: int theta;
5777: char digit[4];
5778: char digitp[25];
5779:
5780: char fileresprobmorprev[FILENAMELENGTH];
5781:
5782: if(popbased==1){
5783: if(mobilav!=0)
5784: strcpy(digitp,"-POPULBASED-MOBILAV_");
5785: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5786: }
5787: else
5788: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5789:
1.218 brouard 5790: /* if (mobilav!=0) { */
5791: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5792: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5793: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5794: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5795: /* } */
5796: /* } */
5797:
5798: strcpy(fileresprobmorprev,"PRMORPREV-");
5799: sprintf(digit,"%-d",ij);
5800: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5801: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5802: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5803: strcat(fileresprobmorprev,fileresu);
5804: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5805: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5806: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5807: }
5808: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5809: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5810: pstamp(ficresprobmorprev);
5811: 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 5812: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5813: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5814: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5815: }
5816: for(j=1;j<=cptcoveff;j++)
5817: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5818: fprintf(ficresprobmorprev,"\n");
5819:
1.218 brouard 5820: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5821: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5822: fprintf(ficresprobmorprev," p.%-d SE",j);
5823: for(i=1; i<=nlstate;i++)
5824: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5825: }
5826: fprintf(ficresprobmorprev,"\n");
5827:
5828: fprintf(ficgp,"\n# Routine varevsij");
5829: fprintf(ficgp,"\nunset title \n");
5830: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5831: 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");
5832: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5833: /* } */
5834: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5835: pstamp(ficresvij);
5836: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5837: if(popbased==1)
5838: 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);
5839: else
5840: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5841: fprintf(ficresvij,"# Age");
5842: for(i=1; i<=nlstate;i++)
5843: for(j=1; j<=nlstate;j++)
5844: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5845: fprintf(ficresvij,"\n");
5846:
5847: xp=vector(1,npar);
5848: dnewm=matrix(1,nlstate,1,npar);
5849: doldm=matrix(1,nlstate,1,nlstate);
5850: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5851: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5852:
5853: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5854: gpp=vector(nlstate+1,nlstate+ndeath);
5855: gmp=vector(nlstate+1,nlstate+ndeath);
5856: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5857:
1.218 brouard 5858: if(estepm < stepm){
5859: printf ("Problem %d lower than %d\n",estepm, stepm);
5860: }
5861: else hstepm=estepm;
5862: /* For example we decided to compute the life expectancy with the smallest unit */
5863: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5864: nhstepm is the number of hstepm from age to agelim
5865: nstepm is the number of stepm from age to agelim.
5866: Look at function hpijx to understand why because of memory size limitations,
5867: we decided (b) to get a life expectancy respecting the most precise curvature of the
5868: survival function given by stepm (the optimization length). Unfortunately it
5869: means that if the survival funtion is printed every two years of age and if
5870: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5871: results. So we changed our mind and took the option of the best precision.
5872: */
5873: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5874: agelim = AGESUP;
5875: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5876: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5877: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5878: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5879: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5880: gp=matrix(0,nhstepm,1,nlstate);
5881: gm=matrix(0,nhstepm,1,nlstate);
5882:
5883:
5884: for(theta=1; theta <=npar; theta++){
5885: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5886: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5887: }
5888:
1.242 brouard 5889: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5890:
5891: if (popbased==1) {
5892: if(mobilav ==0){
5893: for(i=1; i<=nlstate;i++)
5894: prlim[i][i]=probs[(int)age][i][ij];
5895: }else{ /* mobilav */
5896: for(i=1; i<=nlstate;i++)
5897: prlim[i][i]=mobaverage[(int)age][i][ij];
5898: }
5899: }
5900:
1.235 brouard 5901: 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 5902: for(j=1; j<= nlstate; j++){
5903: for(h=0; h<=nhstepm; h++){
5904: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5905: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5906: }
5907: }
5908: /* Next for computing probability of death (h=1 means
5909: computed over hstepm matrices product = hstepm*stepm months)
5910: as a weighted average of prlim.
5911: */
5912: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5913: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5914: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5915: }
5916: /* end probability of death */
5917:
5918: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5919: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5920:
1.242 brouard 5921: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5922:
5923: if (popbased==1) {
5924: if(mobilav ==0){
5925: for(i=1; i<=nlstate;i++)
5926: prlim[i][i]=probs[(int)age][i][ij];
5927: }else{ /* mobilav */
5928: for(i=1; i<=nlstate;i++)
5929: prlim[i][i]=mobaverage[(int)age][i][ij];
5930: }
5931: }
5932:
1.235 brouard 5933: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5934:
5935: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5936: for(h=0; h<=nhstepm; h++){
5937: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5938: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5939: }
5940: }
5941: /* This for computing probability of death (h=1 means
5942: computed over hstepm matrices product = hstepm*stepm months)
5943: as a weighted average of prlim.
5944: */
5945: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5946: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5947: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5948: }
5949: /* end probability of death */
5950:
5951: for(j=1; j<= nlstate; j++) /* vareij */
5952: for(h=0; h<=nhstepm; h++){
5953: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5954: }
5955:
5956: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5957: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5958: }
5959:
5960: } /* End theta */
5961:
5962: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5963:
5964: for(h=0; h<=nhstepm; h++) /* veij */
5965: for(j=1; j<=nlstate;j++)
5966: for(theta=1; theta <=npar; theta++)
5967: trgradg[h][j][theta]=gradg[h][theta][j];
5968:
5969: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5970: for(theta=1; theta <=npar; theta++)
5971: trgradgp[j][theta]=gradgp[theta][j];
5972:
5973:
5974: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5975: for(i=1;i<=nlstate;i++)
5976: for(j=1;j<=nlstate;j++)
5977: vareij[i][j][(int)age] =0.;
5978:
5979: for(h=0;h<=nhstepm;h++){
5980: for(k=0;k<=nhstepm;k++){
5981: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5982: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5983: for(i=1;i<=nlstate;i++)
5984: for(j=1;j<=nlstate;j++)
5985: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5986: }
5987: }
5988:
5989: /* pptj */
5990: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5991: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5992: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5993: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5994: varppt[j][i]=doldmp[j][i];
5995: /* end ppptj */
5996: /* x centered again */
5997:
1.242 brouard 5998: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5999:
6000: if (popbased==1) {
6001: if(mobilav ==0){
6002: for(i=1; i<=nlstate;i++)
6003: prlim[i][i]=probs[(int)age][i][ij];
6004: }else{ /* mobilav */
6005: for(i=1; i<=nlstate;i++)
6006: prlim[i][i]=mobaverage[(int)age][i][ij];
6007: }
6008: }
6009:
6010: /* This for computing probability of death (h=1 means
6011: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6012: as a weighted average of prlim.
6013: */
1.235 brouard 6014: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6015: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6016: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6017: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6018: }
6019: /* end probability of death */
6020:
6021: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6022: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6023: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6024: for(i=1; i<=nlstate;i++){
6025: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6026: }
6027: }
6028: fprintf(ficresprobmorprev,"\n");
6029:
6030: fprintf(ficresvij,"%.0f ",age );
6031: for(i=1; i<=nlstate;i++)
6032: for(j=1; j<=nlstate;j++){
6033: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6034: }
6035: fprintf(ficresvij,"\n");
6036: free_matrix(gp,0,nhstepm,1,nlstate);
6037: free_matrix(gm,0,nhstepm,1,nlstate);
6038: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6039: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6040: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6041: } /* End age */
6042: free_vector(gpp,nlstate+1,nlstate+ndeath);
6043: free_vector(gmp,nlstate+1,nlstate+ndeath);
6044: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6045: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6046: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6047: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6048: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6049: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6050: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6051: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6052: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6053: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6054: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6055: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6056: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6057: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6058: 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);
6059: /* 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 6060: */
1.218 brouard 6061: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6062: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6063:
1.218 brouard 6064: free_vector(xp,1,npar);
6065: free_matrix(doldm,1,nlstate,1,nlstate);
6066: free_matrix(dnewm,1,nlstate,1,npar);
6067: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6068: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6069: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6070: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6071: fclose(ficresprobmorprev);
6072: fflush(ficgp);
6073: fflush(fichtm);
6074: } /* end varevsij */
1.126 brouard 6075:
6076: /************ Variance of prevlim ******************/
1.269 brouard 6077: void varprevlim(char fileresvpl[], FILE *ficresvpl, 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 6078: {
1.205 brouard 6079: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6080: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6081:
1.268 brouard 6082: double **dnewmpar,**doldm;
1.126 brouard 6083: int i, j, nhstepm, hstepm;
6084: double *xp;
6085: double *gp, *gm;
6086: double **gradg, **trgradg;
1.208 brouard 6087: double **mgm, **mgp;
1.126 brouard 6088: double age,agelim;
6089: int theta;
6090:
6091: pstamp(ficresvpl);
6092: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6093: fprintf(ficresvpl,"# Age ");
6094: if(nresult >=1)
6095: fprintf(ficresvpl," Result# ");
1.126 brouard 6096: for(i=1; i<=nlstate;i++)
6097: fprintf(ficresvpl," %1d-%1d",i,i);
6098: fprintf(ficresvpl,"\n");
6099:
6100: xp=vector(1,npar);
1.268 brouard 6101: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6102: doldm=matrix(1,nlstate,1,nlstate);
6103:
6104: hstepm=1*YEARM; /* Every year of age */
6105: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6106: agelim = AGESUP;
6107: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6108: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6109: if (stepm >= YEARM) hstepm=1;
6110: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6111: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6112: mgp=matrix(1,npar,1,nlstate);
6113: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6114: gp=vector(1,nlstate);
6115: gm=vector(1,nlstate);
6116:
6117: for(theta=1; theta <=npar; theta++){
6118: for(i=1; i<=npar; i++){ /* Computes gradient */
6119: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6120: }
1.209 brouard 6121: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6122: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6123: else
1.235 brouard 6124: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6125: for(i=1;i<=nlstate;i++){
1.126 brouard 6126: gp[i] = prlim[i][i];
1.208 brouard 6127: mgp[theta][i] = prlim[i][i];
6128: }
1.126 brouard 6129: for(i=1; i<=npar; i++) /* Computes gradient */
6130: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6131: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6132: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6133: else
1.235 brouard 6134: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6135: for(i=1;i<=nlstate;i++){
1.126 brouard 6136: gm[i] = prlim[i][i];
1.208 brouard 6137: mgm[theta][i] = prlim[i][i];
6138: }
1.126 brouard 6139: for(i=1;i<=nlstate;i++)
6140: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6141: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6142: } /* End theta */
6143:
6144: trgradg =matrix(1,nlstate,1,npar);
6145:
6146: for(j=1; j<=nlstate;j++)
6147: for(theta=1; theta <=npar; theta++)
6148: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6149: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6150: /* printf("\nmgm mgp %d ",(int)age); */
6151: /* for(j=1; j<=nlstate;j++){ */
6152: /* printf(" %d ",j); */
6153: /* for(theta=1; theta <=npar; theta++) */
6154: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6155: /* printf("\n "); */
6156: /* } */
6157: /* } */
6158: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6159: /* printf("\n gradg %d ",(int)age); */
6160: /* for(j=1; j<=nlstate;j++){ */
6161: /* printf("%d ",j); */
6162: /* for(theta=1; theta <=npar; theta++) */
6163: /* printf("%d %lf ",theta,gradg[theta][j]); */
6164: /* printf("\n "); */
6165: /* } */
6166: /* } */
1.126 brouard 6167:
6168: for(i=1;i<=nlstate;i++)
6169: varpl[i][(int)age] =0.;
1.209 brouard 6170: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6171: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6172: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6173: }else{
1.268 brouard 6174: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6175: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6176: }
1.126 brouard 6177: for(i=1;i<=nlstate;i++)
6178: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6179:
6180: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6181: if(nresult >=1)
6182: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6183: for(i=1; i<=nlstate;i++)
6184: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6185: fprintf(ficresvpl,"\n");
6186: free_vector(gp,1,nlstate);
6187: free_vector(gm,1,nlstate);
1.208 brouard 6188: free_matrix(mgm,1,npar,1,nlstate);
6189: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6190: free_matrix(gradg,1,npar,1,nlstate);
6191: free_matrix(trgradg,1,nlstate,1,npar);
6192: } /* End age */
6193:
6194: free_vector(xp,1,npar);
6195: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6196: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6197:
6198: }
6199:
6200:
6201: /************ Variance of backprevalence limit ******************/
1.269 brouard 6202: void varbrevlim(char fileresvbl[], FILE *ficresvbl, double **varbpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **bprlim, double ftolpl, int mobilavproj, int *ncvyearp, int ij, char strstart[], int nres)
1.268 brouard 6203: {
6204: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6205: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6206:
6207: double **dnewmpar,**doldm;
6208: int i, j, nhstepm, hstepm;
6209: double *xp;
6210: double *gp, *gm;
6211: double **gradg, **trgradg;
6212: double **mgm, **mgp;
6213: double age,agelim;
6214: int theta;
6215:
6216: pstamp(ficresvbl);
6217: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6218: fprintf(ficresvbl,"# Age ");
6219: if(nresult >=1)
6220: fprintf(ficresvbl," Result# ");
6221: for(i=1; i<=nlstate;i++)
6222: fprintf(ficresvbl," %1d-%1d",i,i);
6223: fprintf(ficresvbl,"\n");
6224:
6225: xp=vector(1,npar);
6226: dnewmpar=matrix(1,nlstate,1,npar);
6227: doldm=matrix(1,nlstate,1,nlstate);
6228:
6229: hstepm=1*YEARM; /* Every year of age */
6230: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6231: agelim = AGEINF;
6232: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6233: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6234: if (stepm >= YEARM) hstepm=1;
6235: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6236: gradg=matrix(1,npar,1,nlstate);
6237: mgp=matrix(1,npar,1,nlstate);
6238: mgm=matrix(1,npar,1,nlstate);
6239: gp=vector(1,nlstate);
6240: gm=vector(1,nlstate);
6241:
6242: for(theta=1; theta <=npar; theta++){
6243: for(i=1; i<=npar; i++){ /* Computes gradient */
6244: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6245: }
6246: if(mobilavproj > 0 )
6247: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6248: else
6249: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6250: for(i=1;i<=nlstate;i++){
6251: gp[i] = bprlim[i][i];
6252: mgp[theta][i] = bprlim[i][i];
6253: }
6254: for(i=1; i<=npar; i++) /* Computes gradient */
6255: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6256: if(mobilavproj > 0 )
6257: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6258: else
6259: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6260: for(i=1;i<=nlstate;i++){
6261: gm[i] = bprlim[i][i];
6262: mgm[theta][i] = bprlim[i][i];
6263: }
6264: for(i=1;i<=nlstate;i++)
6265: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6266: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6267: } /* End theta */
6268:
6269: trgradg =matrix(1,nlstate,1,npar);
6270:
6271: for(j=1; j<=nlstate;j++)
6272: for(theta=1; theta <=npar; theta++)
6273: trgradg[j][theta]=gradg[theta][j];
6274: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6275: /* printf("\nmgm mgp %d ",(int)age); */
6276: /* for(j=1; j<=nlstate;j++){ */
6277: /* printf(" %d ",j); */
6278: /* for(theta=1; theta <=npar; theta++) */
6279: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6280: /* printf("\n "); */
6281: /* } */
6282: /* } */
6283: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6284: /* printf("\n gradg %d ",(int)age); */
6285: /* for(j=1; j<=nlstate;j++){ */
6286: /* printf("%d ",j); */
6287: /* for(theta=1; theta <=npar; theta++) */
6288: /* printf("%d %lf ",theta,gradg[theta][j]); */
6289: /* printf("\n "); */
6290: /* } */
6291: /* } */
6292:
6293: for(i=1;i<=nlstate;i++)
6294: varbpl[i][(int)age] =0.;
6295: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6296: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6297: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6298: }else{
6299: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6300: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6301: }
6302: for(i=1;i<=nlstate;i++)
6303: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6304:
6305: fprintf(ficresvbl,"%.0f ",age );
6306: if(nresult >=1)
6307: fprintf(ficresvbl,"%d ",nres );
6308: for(i=1; i<=nlstate;i++)
6309: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6310: fprintf(ficresvbl,"\n");
6311: free_vector(gp,1,nlstate);
6312: free_vector(gm,1,nlstate);
6313: free_matrix(mgm,1,npar,1,nlstate);
6314: free_matrix(mgp,1,npar,1,nlstate);
6315: free_matrix(gradg,1,npar,1,nlstate);
6316: free_matrix(trgradg,1,nlstate,1,npar);
6317: } /* End age */
6318:
6319: free_vector(xp,1,npar);
6320: free_matrix(doldm,1,nlstate,1,npar);
6321: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6322:
6323: }
6324:
6325: /************ Variance of one-step probabilities ******************/
6326: 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 6327: {
6328: int i, j=0, k1, l1, tj;
6329: int k2, l2, j1, z1;
6330: int k=0, l;
6331: int first=1, first1, first2;
6332: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6333: double **dnewm,**doldm;
6334: double *xp;
6335: double *gp, *gm;
6336: double **gradg, **trgradg;
6337: double **mu;
6338: double age, cov[NCOVMAX+1];
6339: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6340: int theta;
6341: char fileresprob[FILENAMELENGTH];
6342: char fileresprobcov[FILENAMELENGTH];
6343: char fileresprobcor[FILENAMELENGTH];
6344: double ***varpij;
6345:
6346: strcpy(fileresprob,"PROB_");
6347: strcat(fileresprob,fileres);
6348: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6349: printf("Problem with resultfile: %s\n", fileresprob);
6350: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6351: }
6352: strcpy(fileresprobcov,"PROBCOV_");
6353: strcat(fileresprobcov,fileresu);
6354: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6355: printf("Problem with resultfile: %s\n", fileresprobcov);
6356: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6357: }
6358: strcpy(fileresprobcor,"PROBCOR_");
6359: strcat(fileresprobcor,fileresu);
6360: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6361: printf("Problem with resultfile: %s\n", fileresprobcor);
6362: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6363: }
6364: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6365: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6366: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6367: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6368: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6369: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6370: pstamp(ficresprob);
6371: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6372: fprintf(ficresprob,"# Age");
6373: pstamp(ficresprobcov);
6374: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6375: fprintf(ficresprobcov,"# Age");
6376: pstamp(ficresprobcor);
6377: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6378: fprintf(ficresprobcor,"# Age");
1.126 brouard 6379:
6380:
1.222 brouard 6381: for(i=1; i<=nlstate;i++)
6382: for(j=1; j<=(nlstate+ndeath);j++){
6383: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6384: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6385: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6386: }
6387: /* fprintf(ficresprob,"\n");
6388: fprintf(ficresprobcov,"\n");
6389: fprintf(ficresprobcor,"\n");
6390: */
6391: xp=vector(1,npar);
6392: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6393: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6394: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6395: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6396: first=1;
6397: fprintf(ficgp,"\n# Routine varprob");
6398: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6399: fprintf(fichtm,"\n");
6400:
1.266 brouard 6401: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back. %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6402: 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);
6403: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6404: and drawn. It helps understanding how is the covariance between two incidences.\
6405: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6406: 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 6407: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6408: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6409: standard deviations wide on each axis. <br>\
6410: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6411: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6412: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6413:
1.222 brouard 6414: cov[1]=1;
6415: /* tj=cptcoveff; */
1.225 brouard 6416: tj = (int) pow(2,cptcoveff);
1.222 brouard 6417: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6418: j1=0;
1.224 brouard 6419: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6420: if (cptcovn>0) {
6421: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6422: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6423: fprintf(ficresprob, "**********\n#\n");
6424: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6425: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6426: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6427:
1.222 brouard 6428: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6429: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6430: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6431:
6432:
1.222 brouard 6433: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6434: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6435: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6436:
1.222 brouard 6437: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6438: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6439: fprintf(ficresprobcor, "**********\n#");
6440: if(invalidvarcomb[j1]){
6441: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6442: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6443: continue;
6444: }
6445: }
6446: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6447: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6448: gp=vector(1,(nlstate)*(nlstate+ndeath));
6449: gm=vector(1,(nlstate)*(nlstate+ndeath));
6450: for (age=bage; age<=fage; age ++){
6451: cov[2]=age;
6452: if(nagesqr==1)
6453: cov[3]= age*age;
6454: for (k=1; k<=cptcovn;k++) {
6455: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6456: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6457: * 1 1 1 1 1
6458: * 2 2 1 1 1
6459: * 3 1 2 1 1
6460: */
6461: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6462: }
6463: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6464: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6465: for (k=1; k<=cptcovprod;k++)
6466: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6467:
6468:
1.222 brouard 6469: for(theta=1; theta <=npar; theta++){
6470: for(i=1; i<=npar; i++)
6471: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6472:
1.222 brouard 6473: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6474:
1.222 brouard 6475: k=0;
6476: for(i=1; i<= (nlstate); i++){
6477: for(j=1; j<=(nlstate+ndeath);j++){
6478: k=k+1;
6479: gp[k]=pmmij[i][j];
6480: }
6481: }
1.220 brouard 6482:
1.222 brouard 6483: for(i=1; i<=npar; i++)
6484: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6485:
1.222 brouard 6486: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6487: k=0;
6488: for(i=1; i<=(nlstate); i++){
6489: for(j=1; j<=(nlstate+ndeath);j++){
6490: k=k+1;
6491: gm[k]=pmmij[i][j];
6492: }
6493: }
1.220 brouard 6494:
1.222 brouard 6495: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6496: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6497: }
1.126 brouard 6498:
1.222 brouard 6499: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6500: for(theta=1; theta <=npar; theta++)
6501: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6502:
1.222 brouard 6503: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6504: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6505:
1.222 brouard 6506: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6507:
1.222 brouard 6508: k=0;
6509: for(i=1; i<=(nlstate); i++){
6510: for(j=1; j<=(nlstate+ndeath);j++){
6511: k=k+1;
6512: mu[k][(int) age]=pmmij[i][j];
6513: }
6514: }
6515: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6516: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6517: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6518:
1.222 brouard 6519: /*printf("\n%d ",(int)age);
6520: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6521: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6522: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6523: }*/
1.220 brouard 6524:
1.222 brouard 6525: fprintf(ficresprob,"\n%d ",(int)age);
6526: fprintf(ficresprobcov,"\n%d ",(int)age);
6527: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6528:
1.222 brouard 6529: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6530: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6531: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6532: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6533: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6534: }
6535: i=0;
6536: for (k=1; k<=(nlstate);k++){
6537: for (l=1; l<=(nlstate+ndeath);l++){
6538: i++;
6539: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6540: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6541: for (j=1; j<=i;j++){
6542: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6543: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6544: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6545: }
6546: }
6547: }/* end of loop for state */
6548: } /* end of loop for age */
6549: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6550: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6551: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6552: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6553:
6554: /* Confidence intervalle of pij */
6555: /*
6556: fprintf(ficgp,"\nunset parametric;unset label");
6557: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6558: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6559: 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);
6560: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6561: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6562: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6563: */
6564:
6565: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6566: first1=1;first2=2;
6567: for (k2=1; k2<=(nlstate);k2++){
6568: for (l2=1; l2<=(nlstate+ndeath);l2++){
6569: if(l2==k2) continue;
6570: j=(k2-1)*(nlstate+ndeath)+l2;
6571: for (k1=1; k1<=(nlstate);k1++){
6572: for (l1=1; l1<=(nlstate+ndeath);l1++){
6573: if(l1==k1) continue;
6574: i=(k1-1)*(nlstate+ndeath)+l1;
6575: if(i<=j) continue;
6576: for (age=bage; age<=fage; age ++){
6577: if ((int)age %5==0){
6578: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6579: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6580: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6581: mu1=mu[i][(int) age]/stepm*YEARM ;
6582: mu2=mu[j][(int) age]/stepm*YEARM;
6583: c12=cv12/sqrt(v1*v2);
6584: /* Computing eigen value of matrix of covariance */
6585: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6586: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6587: if ((lc2 <0) || (lc1 <0) ){
6588: if(first2==1){
6589: first1=0;
6590: 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);
6591: }
6592: 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);
6593: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6594: /* lc2=fabs(lc2); */
6595: }
1.220 brouard 6596:
1.222 brouard 6597: /* Eigen vectors */
6598: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6599: /*v21=sqrt(1.-v11*v11); *//* error */
6600: v21=(lc1-v1)/cv12*v11;
6601: v12=-v21;
6602: v22=v11;
6603: tnalp=v21/v11;
6604: if(first1==1){
6605: first1=0;
6606: 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);
6607: }
6608: 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);
6609: /*printf(fignu*/
6610: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6611: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6612: if(first==1){
6613: first=0;
6614: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6615: fprintf(ficgp,"\nset parametric;unset label");
6616: 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);
6617: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6618: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6619: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6620: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6621: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6622: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6623: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6624: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6625: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6626: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6627: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6628: fprintf(ficgp,"\nplot [-pi:pi] %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.266 brouard 6629: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6630: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6631: }else{
6632: first=0;
6633: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6634: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6635: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6636: fprintf(ficgp,"\nreplot %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.266 brouard 6637: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6638: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6639: }/* if first */
6640: } /* age mod 5 */
6641: } /* end loop age */
6642: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6643: first=1;
6644: } /*l12 */
6645: } /* k12 */
6646: } /*l1 */
6647: }/* k1 */
6648: } /* loop on combination of covariates j1 */
6649: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6650: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6651: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6652: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6653: free_vector(xp,1,npar);
6654: fclose(ficresprob);
6655: fclose(ficresprobcov);
6656: fclose(ficresprobcor);
6657: fflush(ficgp);
6658: fflush(fichtmcov);
6659: }
1.126 brouard 6660:
6661:
6662: /******************* Printing html file ***********/
1.201 brouard 6663: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6664: int lastpass, int stepm, int weightopt, char model[],\
6665: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6666: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 ! brouard 6667: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
! 6668: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6669: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6670:
6671: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6672: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6673: </ul>");
1.237 brouard 6674: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6675: </ul>", model);
1.214 brouard 6676: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6677: 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",
6678: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6679: 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 6680: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6681: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6682: fprintf(fichtm,"\
6683: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6684: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6685: fprintf(fichtm,"\
1.217 brouard 6686: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6687: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6688: fprintf(fichtm,"\
1.126 brouard 6689: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6690: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6691: fprintf(fichtm,"\
1.217 brouard 6692: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6693: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6694: fprintf(fichtm,"\
1.211 brouard 6695: - (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 6696: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6697: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6698: if(prevfcast==1){
6699: fprintf(fichtm,"\
6700: - Prevalence projections by age and states: \
1.201 brouard 6701: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6702: }
1.126 brouard 6703:
6704:
1.225 brouard 6705: m=pow(2,cptcoveff);
1.222 brouard 6706: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6707:
1.264 brouard 6708: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6709:
6710: jj1=0;
6711:
6712: fprintf(fichtm," \n<ul>");
6713: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6714: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6715: if(m != 1 && TKresult[nres]!= k1)
6716: continue;
6717: jj1++;
6718: if (cptcovn > 0) {
6719: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6720: for (cpt=1; cpt<=cptcoveff;cpt++){
6721: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6722: }
6723: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6724: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6725: }
6726: fprintf(fichtm,"\">");
6727:
6728: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6729: fprintf(fichtm,"************ Results for covariates");
6730: for (cpt=1; cpt<=cptcoveff;cpt++){
6731: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6732: }
6733: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6734: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6735: }
6736: if(invalidvarcomb[k1]){
6737: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6738: continue;
6739: }
6740: fprintf(fichtm,"</a></li>");
6741: } /* cptcovn >0 */
6742: }
6743: fprintf(fichtm," \n</ul>");
6744:
1.222 brouard 6745: jj1=0;
1.237 brouard 6746:
6747: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6748: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6749: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6750: continue;
1.220 brouard 6751:
1.222 brouard 6752: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6753: jj1++;
6754: if (cptcovn > 0) {
1.264 brouard 6755: fprintf(fichtm,"\n<p><a name=\"rescov");
6756: for (cpt=1; cpt<=cptcoveff;cpt++){
6757: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6758: }
6759: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6760: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6761: }
6762: fprintf(fichtm,"\"</a>");
6763:
1.222 brouard 6764: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6765: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6766: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6767: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6768: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6769: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6770: }
1.237 brouard 6771: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6772: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6773: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6774: }
6775:
1.230 brouard 6776: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6777: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6778: if(invalidvarcomb[k1]){
6779: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6780: printf("\nCombination (%d) ignored because no cases \n",k1);
6781: continue;
6782: }
6783: }
6784: /* aij, bij */
1.259 brouard 6785: 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 6786: <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 6787: /* Pij */
1.241 brouard 6788: 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> \
6789: <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 6790: /* Quasi-incidences */
6791: 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 6792: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6793: 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 6794: 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> \
6795: <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 6796: /* Survival functions (period) in state j */
6797: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6798: 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> \
6799: <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 6800: }
6801: /* State specific survival functions (period) */
6802: for(cpt=1; cpt<=nlstate;cpt++){
6803: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6804: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6805: <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 6806: }
6807: /* Period (stable) prevalence in each health state */
6808: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6809: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6810: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 6811: }
6812: if(backcast==1){
6813: /* Period (stable) back prevalence in each health state */
6814: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6815: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6816: <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 6817: }
1.217 brouard 6818: }
1.222 brouard 6819: if(prevfcast==1){
6820: /* Projection of prevalence up to period (stable) prevalence in each health state */
6821: for(cpt=1; cpt<=nlstate;cpt++){
1.273 ! brouard 6822: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
! 6823: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateproj1, dateproj2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6824: }
6825: }
1.268 brouard 6826: if(backcast==1){
6827: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6828: for(cpt=1; cpt<=nlstate;cpt++){
1.273 ! brouard 6829: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
! 6830: from year %.1f up to year %.1f (probably close to stable [mixed] back prevalence in state %d (randomness in cross-sectional prevalence is not taken into \
! 6831: account but can visually be appreciated). Or probability to have been in an state %d, knowing that the person was in either state (1 or %d) \
! 6832: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
! 6833: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 6834: }
6835: }
1.220 brouard 6836:
1.222 brouard 6837: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6838: 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> \
6839: <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 6840: }
6841: /* } /\* end i1 *\/ */
6842: }/* End k1 */
6843: fprintf(fichtm,"</ul>");
1.126 brouard 6844:
1.222 brouard 6845: fprintf(fichtm,"\
1.126 brouard 6846: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6847: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6848: - 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 6849: But because parameters are usually highly correlated (a higher incidence of disability \
6850: and a higher incidence of recovery can give very close observed transition) it might \
6851: be very useful to look not only at linear confidence intervals estimated from the \
6852: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6853: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6854: covariance matrix of the one-step probabilities. \
6855: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6856:
1.222 brouard 6857: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6858: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6859: fprintf(fichtm,"\
1.126 brouard 6860: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6861: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6862:
1.222 brouard 6863: fprintf(fichtm,"\
1.126 brouard 6864: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6865: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6866: fprintf(fichtm,"\
1.126 brouard 6867: - 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): \
6868: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6869: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6870: fprintf(fichtm,"\
1.126 brouard 6871: - (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): \
6872: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6873: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6874: fprintf(fichtm,"\
1.128 brouard 6875: - 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 6876: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6877: fprintf(fichtm,"\
1.128 brouard 6878: - 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 6879: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6880: fprintf(fichtm,"\
1.126 brouard 6881: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6882: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6883:
6884: /* if(popforecast==1) fprintf(fichtm,"\n */
6885: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6886: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6887: /* <br>",fileres,fileres,fileres,fileres); */
6888: /* else */
6889: /* 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 6890: fflush(fichtm);
6891: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6892:
1.225 brouard 6893: m=pow(2,cptcoveff);
1.222 brouard 6894: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6895:
1.222 brouard 6896: jj1=0;
1.237 brouard 6897:
1.241 brouard 6898: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6899: for(k1=1; k1<=m;k1++){
1.253 brouard 6900: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6901: continue;
1.222 brouard 6902: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6903: jj1++;
1.126 brouard 6904: if (cptcovn > 0) {
6905: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6906: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6907: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6908: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6909: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6910: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6911: }
6912:
1.126 brouard 6913: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6914:
1.222 brouard 6915: if(invalidvarcomb[k1]){
6916: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6917: continue;
6918: }
1.126 brouard 6919: }
6920: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6921: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6922: 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 6923: <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 6924: }
6925: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6926: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6927: true period expectancies (those weighted with period prevalences are also\
6928: drawn in addition to the population based expectancies computed using\
1.241 brouard 6929: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6930: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6931: /* } /\* end i1 *\/ */
6932: }/* End k1 */
1.241 brouard 6933: }/* End nres */
1.222 brouard 6934: fprintf(fichtm,"</ul>");
6935: fflush(fichtm);
1.126 brouard 6936: }
6937:
6938: /******************* Gnuplot file **************/
1.270 brouard 6939: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 6940:
6941: char dirfileres[132],optfileres[132];
1.264 brouard 6942: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6943: 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 6944: int lv=0, vlv=0, kl=0;
1.130 brouard 6945: int ng=0;
1.201 brouard 6946: int vpopbased;
1.223 brouard 6947: int ioffset; /* variable offset for columns */
1.270 brouard 6948: int iyearc=1; /* variable column for year of projection */
6949: int iagec=1; /* variable column for age of projection */
1.235 brouard 6950: int nres=0; /* Index of resultline */
1.266 brouard 6951: int istart=1; /* For starting graphs in projections */
1.219 brouard 6952:
1.126 brouard 6953: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6954: /* printf("Problem with file %s",optionfilegnuplot); */
6955: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6956: /* } */
6957:
6958: /*#ifdef windows */
6959: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6960: /*#endif */
1.225 brouard 6961: m=pow(2,cptcoveff);
1.126 brouard 6962:
1.202 brouard 6963: /* Contribution to likelihood */
6964: /* Plot the probability implied in the likelihood */
1.223 brouard 6965: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6966: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6967: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6968: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6969: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6970: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6971: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6972: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6973: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6974: 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));
6975: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6976: 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));
6977: for (i=1; i<= nlstate ; i ++) {
6978: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6979: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6980: 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);
6981: for (j=2; j<= nlstate+ndeath ; j ++) {
6982: 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);
6983: }
6984: fprintf(ficgp,";\nset out; unset ylabel;\n");
6985: }
6986: /* 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 */
6987: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6988: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6989: fprintf(ficgp,"\nset out;unset log\n");
6990: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6991:
1.126 brouard 6992: strcpy(dirfileres,optionfilefiname);
6993: strcpy(optfileres,"vpl");
1.223 brouard 6994: /* 1eme*/
1.238 brouard 6995: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6996: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6997: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6998: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6999: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7000: continue;
7001: /* We are interested in selected combination by the resultline */
1.246 brouard 7002: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7003: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7004: strcpy(gplotlabel,"(");
1.238 brouard 7005: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7006: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7007: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7008: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7009: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7010: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7011: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7012: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7013: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7014: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7015: }
7016: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7017: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7018: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7019: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7020: }
7021: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7022: /* printf("\n#\n"); */
1.238 brouard 7023: fprintf(ficgp,"\n#\n");
7024: if(invalidvarcomb[k1]){
1.260 brouard 7025: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7026: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7027: continue;
7028: }
1.235 brouard 7029:
1.241 brouard 7030: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7031: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.264 brouard 7032: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7033: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
7034: /* 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); */
7035: /* k1-1 error should be nres-1*/
1.238 brouard 7036: for (i=1; i<= nlstate ; i ++) {
7037: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7038: else fprintf(ficgp," %%*lf (%%*lf)");
7039: }
1.260 brouard 7040: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7041: for (i=1; i<= nlstate ; i ++) {
7042: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7043: else fprintf(ficgp," %%*lf (%%*lf)");
7044: }
1.260 brouard 7045: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7046: for (i=1; i<= nlstate ; i ++) {
7047: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7048: else fprintf(ficgp," %%*lf (%%*lf)");
7049: }
1.265 brouard 7050: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7051:
7052: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7053: if(cptcoveff ==0){
1.271 brouard 7054: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7055: }else{
7056: kl=0;
7057: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7058: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7059: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7060: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7061: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7062: vlv= nbcode[Tvaraff[k]][lv];
7063: kl++;
7064: /* 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 *\/ */
7065: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7066: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7067: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7068: if(k==cptcoveff){
7069: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7070: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7071: }else{
7072: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7073: kl++;
7074: }
7075: } /* end covariate */
7076: } /* end if no covariate */
7077:
1.238 brouard 7078: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7079: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7080: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7081: if(cptcoveff ==0){
1.245 brouard 7082: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7083: }else{
7084: kl=0;
7085: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7086: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7087: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7088: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7089: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7090: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7091: kl++;
1.238 brouard 7092: /* 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 *\/ */
7093: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7094: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7095: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7096: if(k==cptcoveff){
1.245 brouard 7097: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7098: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7099: }else{
7100: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7101: kl++;
7102: }
7103: } /* end covariate */
7104: } /* end if no covariate */
1.268 brouard 7105: if(backcast == 1){
7106: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7107: /* k1-1 error should be nres-1*/
7108: for (i=1; i<= nlstate ; i ++) {
7109: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7110: else fprintf(ficgp," %%*lf (%%*lf)");
7111: }
1.271 brouard 7112: fprintf(ficgp,"\" t\"Backward (stable) prevalence\" w l lt 6 dt 3,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 7113: for (i=1; i<= nlstate ; i ++) {
7114: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7115: else fprintf(ficgp," %%*lf (%%*lf)");
7116: }
1.272 brouard 7117: fprintf(ficgp,"\" t\"95%% CI\" w l lt 5,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 7118: for (i=1; i<= nlstate ; i ++) {
7119: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7120: else fprintf(ficgp," %%*lf (%%*lf)");
7121: }
1.272 brouard 7122: fprintf(ficgp,"\" t\"\" w l lt 5");
1.268 brouard 7123: } /* end if backprojcast */
1.238 brouard 7124: } /* end if backcast */
1.264 brouard 7125: fprintf(ficgp,"\nset out ;unset label;\n");
1.238 brouard 7126: } /* nres */
1.201 brouard 7127: } /* k1 */
7128: } /* cpt */
1.235 brouard 7129:
7130:
1.126 brouard 7131: /*2 eme*/
1.238 brouard 7132: for (k1=1; k1<= m ; k1 ++){
7133: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7134: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7135: continue;
7136: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7137: strcpy(gplotlabel,"(");
1.238 brouard 7138: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7139: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7140: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7141: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7142: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7143: vlv= nbcode[Tvaraff[k]][lv];
7144: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7145: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7146: }
1.237 brouard 7147: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7148: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7149: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7150: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7151: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7152: }
1.264 brouard 7153: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7154: fprintf(ficgp,"\n#\n");
1.223 brouard 7155: if(invalidvarcomb[k1]){
7156: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7157: continue;
7158: }
1.219 brouard 7159:
1.241 brouard 7160: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7161: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7162: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7163: if(vpopbased==0){
1.238 brouard 7164: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7165: }else
1.238 brouard 7166: fprintf(ficgp,"\nreplot ");
7167: for (i=1; i<= nlstate+1 ; i ++) {
7168: k=2*i;
1.261 brouard 7169: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased);
1.238 brouard 7170: for (j=1; j<= nlstate+1 ; j ++) {
7171: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7172: else fprintf(ficgp," %%*lf (%%*lf)");
7173: }
7174: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7175: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7176: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 7177: for (j=1; j<= nlstate+1 ; j ++) {
7178: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7179: else fprintf(ficgp," %%*lf (%%*lf)");
7180: }
7181: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7182: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 7183: for (j=1; j<= nlstate+1 ; j ++) {
7184: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7185: else fprintf(ficgp," %%*lf (%%*lf)");
7186: }
7187: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7188: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7189: } /* state */
7190: } /* vpopbased */
1.264 brouard 7191: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; unset label;\n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 7192: } /* end nres */
7193: } /* k1 end 2 eme*/
7194:
7195:
7196: /*3eme*/
7197: for (k1=1; k1<= m ; k1 ++){
7198: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7199: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7200: continue;
7201:
7202: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7203: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7204: strcpy(gplotlabel,"(");
1.238 brouard 7205: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7206: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7207: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7208: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7209: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7210: vlv= nbcode[Tvaraff[k]][lv];
7211: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7212: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7213: }
7214: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7215: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7216: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7217: }
1.264 brouard 7218: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7219: fprintf(ficgp,"\n#\n");
7220: if(invalidvarcomb[k1]){
7221: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7222: continue;
7223: }
7224:
7225: /* k=2+nlstate*(2*cpt-2); */
7226: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7227: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7228: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7229: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7230: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),nres-1,nres-1,k,cpt);
1.238 brouard 7231: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7232: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7233: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7234: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7235: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7236: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7237:
1.238 brouard 7238: */
7239: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7240: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+i,cpt,i+1);
1.238 brouard 7241: /* 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 7242:
1.238 brouard 7243: }
1.261 brouard 7244: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+nlstate,cpt);
1.238 brouard 7245: }
1.264 brouard 7246: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7247: } /* end nres */
7248: } /* end kl 3eme */
1.126 brouard 7249:
1.223 brouard 7250: /* 4eme */
1.201 brouard 7251: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7252: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7253: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7254: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7255: continue;
1.238 brouard 7256: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7257: strcpy(gplotlabel,"(");
1.238 brouard 7258: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7259: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7260: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7261: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7262: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7263: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7264: vlv= nbcode[Tvaraff[k]][lv];
7265: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7266: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7267: }
7268: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7269: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7270: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7271: }
1.264 brouard 7272: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7273: fprintf(ficgp,"\n#\n");
7274: if(invalidvarcomb[k1]){
7275: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7276: continue;
1.223 brouard 7277: }
1.238 brouard 7278:
1.241 brouard 7279: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7280: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 7281: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7282: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7283: k=3;
7284: for (i=1; i<= nlstate ; i ++){
7285: if(i==1){
7286: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7287: }else{
7288: fprintf(ficgp,", '' ");
7289: }
7290: l=(nlstate+ndeath)*(i-1)+1;
7291: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7292: for (j=2; j<= nlstate+ndeath ; j ++)
7293: fprintf(ficgp,"+$%d",k+l+j-1);
7294: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7295: } /* nlstate */
1.264 brouard 7296: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7297: } /* end cpt state*/
7298: } /* end nres */
7299: } /* end covariate k1 */
7300:
1.220 brouard 7301: /* 5eme */
1.201 brouard 7302: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7303: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7304: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7305: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7306: continue;
1.238 brouard 7307: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7308: strcpy(gplotlabel,"(");
1.238 brouard 7309: 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);
7310: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7311: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7312: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7313: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7314: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7315: vlv= nbcode[Tvaraff[k]][lv];
7316: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7317: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7318: }
7319: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7320: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7321: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7322: }
1.264 brouard 7323: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7324: fprintf(ficgp,"\n#\n");
7325: if(invalidvarcomb[k1]){
7326: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7327: continue;
7328: }
1.227 brouard 7329:
1.241 brouard 7330: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7331: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 7332: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7333: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7334: k=3;
7335: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7336: if(j==1)
7337: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7338: else
7339: fprintf(ficgp,", '' ");
7340: l=(nlstate+ndeath)*(cpt-1) +j;
7341: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7342: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7343: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7344: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7345: } /* nlstate */
7346: fprintf(ficgp,", '' ");
7347: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7348: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7349: l=(nlstate+ndeath)*(cpt-1) +j;
7350: if(j < nlstate)
7351: fprintf(ficgp,"$%d +",k+l);
7352: else
7353: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7354: }
1.264 brouard 7355: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7356: } /* end cpt state*/
7357: } /* end covariate */
7358: } /* end nres */
1.227 brouard 7359:
1.220 brouard 7360: /* 6eme */
1.202 brouard 7361: /* CV preval stable (period) for each covariate */
1.237 brouard 7362: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7363: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7364: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7365: continue;
1.255 brouard 7366: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7367: strcpy(gplotlabel,"(");
1.211 brouard 7368: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7369: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7370: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7371: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7372: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7373: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7374: vlv= nbcode[Tvaraff[k]][lv];
7375: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7376: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7377: }
1.237 brouard 7378: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7379: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7380: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7381: }
1.264 brouard 7382: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7383: fprintf(ficgp,"\n#\n");
1.223 brouard 7384: if(invalidvarcomb[k1]){
1.227 brouard 7385: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7386: continue;
1.223 brouard 7387: }
1.227 brouard 7388:
1.241 brouard 7389: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7390: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.126 brouard 7391: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7392: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7393: k=3; /* Offset */
1.255 brouard 7394: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7395: if(i==1)
7396: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7397: else
7398: fprintf(ficgp,", '' ");
1.255 brouard 7399: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7400: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7401: for (j=2; j<= nlstate ; j ++)
7402: fprintf(ficgp,"+$%d",k+l+j-1);
7403: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7404: } /* nlstate */
1.264 brouard 7405: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7406: } /* end cpt state*/
7407: } /* end covariate */
1.227 brouard 7408:
7409:
1.220 brouard 7410: /* 7eme */
1.218 brouard 7411: if(backcast == 1){
1.217 brouard 7412: /* CV back preval stable (period) for each covariate */
1.237 brouard 7413: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7414: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7415: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7416: continue;
1.268 brouard 7417: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7418: strcpy(gplotlabel,"(");
7419: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7420: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7421: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7422: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7423: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7424: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7425: vlv= nbcode[Tvaraff[k]][lv];
7426: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7427: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7428: }
1.237 brouard 7429: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7430: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7431: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7432: }
1.264 brouard 7433: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7434: fprintf(ficgp,"\n#\n");
7435: if(invalidvarcomb[k1]){
7436: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7437: continue;
7438: }
7439:
1.241 brouard 7440: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7441: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7442: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7443: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7444: k=3; /* Offset */
1.268 brouard 7445: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7446: if(i==1)
7447: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7448: else
7449: fprintf(ficgp,", '' ");
7450: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7451: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7452: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7453: /* 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 7454: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7455: /* for (j=2; j<= nlstate ; j ++) */
7456: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7457: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7458: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7459: } /* nlstate */
1.264 brouard 7460: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7461: } /* end cpt state*/
7462: } /* end covariate */
7463: } /* End if backcast */
7464:
1.223 brouard 7465: /* 8eme */
1.218 brouard 7466: if(prevfcast==1){
7467: /* Projection from cross-sectional to stable (period) for each covariate */
7468:
1.237 brouard 7469: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7470: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7471: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7472: continue;
1.211 brouard 7473: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7474: strcpy(gplotlabel,"(");
1.227 brouard 7475: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7476: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7477: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7478: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7479: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7480: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7481: vlv= nbcode[Tvaraff[k]][lv];
7482: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7483: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7484: }
1.237 brouard 7485: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7486: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7487: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7488: }
1.264 brouard 7489: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7490: fprintf(ficgp,"\n#\n");
7491: if(invalidvarcomb[k1]){
7492: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7493: continue;
7494: }
7495:
7496: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7497: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7498: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7499: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7500: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7501:
7502: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7503: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7504: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7505: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7506: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7507: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7508: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7509: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7510: if(i==istart){
1.227 brouard 7511: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7512: }else{
7513: fprintf(ficgp,",\\\n '' ");
7514: }
7515: if(cptcoveff ==0){ /* No covariate */
7516: ioffset=2; /* Age is in 2 */
7517: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7518: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7519: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7520: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7521: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7522: if(i==nlstate+1){
1.270 brouard 7523: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7524: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7525: fprintf(ficgp,",\\\n '' ");
7526: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7527: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7528: offyear, \
1.268 brouard 7529: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7530: }else
1.227 brouard 7531: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7532: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7533: }else{ /* more than 2 covariates */
1.270 brouard 7534: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7535: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7536: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7537: iyearc=ioffset-1;
7538: iagec=ioffset;
1.227 brouard 7539: fprintf(ficgp," u %d:(",ioffset);
7540: kl=0;
7541: strcpy(gplotcondition,"(");
7542: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7543: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7544: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7545: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7546: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7547: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7548: kl++;
7549: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7550: kl++;
7551: if(k <cptcoveff && cptcoveff>1)
7552: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7553: }
7554: strcpy(gplotcondition+strlen(gplotcondition),")");
7555: /* 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 *\/ */
7556: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7557: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7558: /* '' 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*/
7559: if(i==nlstate+1){
1.270 brouard 7560: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7561: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7562: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7563: fprintf(ficgp," u %d:(",iagec);
7564: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7565: iyearc, iagec, offyear, \
7566: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7567: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
1.227 brouard 7568: }else{
7569: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7570: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7571: }
7572: } /* end if covariate */
7573: } /* nlstate */
1.264 brouard 7574: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7575: } /* end cpt state*/
7576: } /* end covariate */
7577: } /* End if prevfcast */
1.227 brouard 7578:
1.268 brouard 7579: if(backcast==1){
7580: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7581:
7582: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7583: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7584: if(m != 1 && TKresult[nres]!= k1)
7585: continue;
7586: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7587: strcpy(gplotlabel,"(");
7588: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7589: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7590: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7591: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7592: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7593: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7594: vlv= nbcode[Tvaraff[k]][lv];
7595: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7596: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7597: }
7598: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7599: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7600: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7601: }
7602: strcpy(gplotlabel+strlen(gplotlabel),")");
7603: fprintf(ficgp,"\n#\n");
7604: if(invalidvarcomb[k1]){
7605: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7606: continue;
7607: }
7608:
7609: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7610: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7611: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7612: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7613: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7614:
7615: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7616: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7617: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7618: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7619: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7620: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7621: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7622: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7623: if(i==istart){
7624: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7625: }else{
7626: fprintf(ficgp,",\\\n '' ");
7627: }
7628: if(cptcoveff ==0){ /* No covariate */
7629: ioffset=2; /* Age is in 2 */
7630: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7631: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7632: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7633: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7634: fprintf(ficgp," u %d:(", ioffset);
7635: if(i==nlstate+1){
1.270 brouard 7636: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7637: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7638: fprintf(ficgp,",\\\n '' ");
7639: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7640: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7641: offbyear, \
7642: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7643: }else
7644: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7645: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7646: }else{ /* more than 2 covariates */
1.270 brouard 7647: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7648: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7649: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7650: iyearc=ioffset-1;
7651: iagec=ioffset;
1.268 brouard 7652: fprintf(ficgp," u %d:(",ioffset);
7653: kl=0;
7654: strcpy(gplotcondition,"(");
7655: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7656: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7657: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7658: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7659: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7660: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7661: kl++;
7662: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7663: kl++;
7664: if(k <cptcoveff && cptcoveff>1)
7665: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7666: }
7667: strcpy(gplotcondition+strlen(gplotcondition),")");
7668: /* 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 *\/ */
7669: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7670: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7671: /* '' 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*/
7672: if(i==nlstate+1){
1.270 brouard 7673: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7674: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7675: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7676: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7677: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7678: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7679: iyearc,iagec,offbyear, \
7680: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7681: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7682: }else{
7683: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7684: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7685: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7686: }
7687: } /* end if covariate */
7688: } /* nlstate */
7689: fprintf(ficgp,"\nset out; unset label;\n");
7690: } /* end cpt state*/
7691: } /* end covariate */
7692: } /* End if backcast */
7693:
1.227 brouard 7694:
1.238 brouard 7695: /* 9eme writing MLE parameters */
7696: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7697: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7698: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7699: for(k=1; k <=(nlstate+ndeath); k++){
7700: if (k != i) {
1.227 brouard 7701: fprintf(ficgp,"# current state %d\n",k);
7702: for(j=1; j <=ncovmodel; j++){
7703: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7704: jk++;
7705: }
7706: fprintf(ficgp,"\n");
1.126 brouard 7707: }
7708: }
1.223 brouard 7709: }
1.187 brouard 7710: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7711:
1.145 brouard 7712: /*goto avoid;*/
1.238 brouard 7713: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7714: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7715: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7716: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7717: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7718: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7719: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7720: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7721: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7722: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7723: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7724: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7725: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7726: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7727: fprintf(ficgp,"#\n");
1.223 brouard 7728: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7729: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7730: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7731: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7732: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7733: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7734: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7735: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7736: continue;
1.264 brouard 7737: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7738: strcpy(gplotlabel,"(");
7739: sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);
7740: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7741: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7742: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7743: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7744: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7745: vlv= nbcode[Tvaraff[k]][lv];
7746: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7747: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7748: }
1.237 brouard 7749: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7750: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7751: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7752: }
1.264 brouard 7753: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7754: fprintf(ficgp,"\n#\n");
1.264 brouard 7755: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
7756: fprintf(ficgp,"\nset label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7757: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7758: if (ng==1){
7759: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7760: fprintf(ficgp,"\nunset log y");
7761: }else if (ng==2){
7762: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7763: fprintf(ficgp,"\nset log y");
7764: }else if (ng==3){
7765: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7766: fprintf(ficgp,"\nset log y");
7767: }else
7768: fprintf(ficgp,"\nunset title ");
7769: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7770: i=1;
7771: for(k2=1; k2<=nlstate; k2++) {
7772: k3=i;
7773: for(k=1; k<=(nlstate+ndeath); k++) {
7774: if (k != k2){
7775: switch( ng) {
7776: case 1:
7777: if(nagesqr==0)
7778: fprintf(ficgp," p%d+p%d*x",i,i+1);
7779: else /* nagesqr =1 */
7780: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7781: break;
7782: case 2: /* ng=2 */
7783: if(nagesqr==0)
7784: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7785: else /* nagesqr =1 */
7786: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7787: break;
7788: case 3:
7789: if(nagesqr==0)
7790: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7791: else /* nagesqr =1 */
7792: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7793: break;
7794: }
7795: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7796: ijp=1; /* product no age */
7797: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7798: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7799: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7800: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7801: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7802: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7803: if(DummyV[j]==0){
7804: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7805: }else{ /* quantitative */
7806: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7807: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7808: }
7809: ij++;
1.237 brouard 7810: }
1.268 brouard 7811: }
7812: }else if(cptcovprod >0){
7813: if(j==Tprod[ijp]) { /* */
7814: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7815: if(ijp <=cptcovprod) { /* Product */
7816: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7817: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7818: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
7819: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7820: }else{ /* Vn is dummy and Vm is quanti */
7821: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7822: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7823: }
7824: }else{ /* Vn*Vm Vn is quanti */
7825: if(DummyV[Tvard[ijp][2]]==0){
7826: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7827: }else{ /* Both quanti */
7828: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7829: }
1.237 brouard 7830: }
1.268 brouard 7831: ijp++;
1.237 brouard 7832: }
1.268 brouard 7833: } /* end Tprod */
1.237 brouard 7834: } else{ /* simple covariate */
1.264 brouard 7835: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7836: if(Dummy[j]==0){
7837: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7838: }else{ /* quantitative */
7839: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7840: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7841: }
1.237 brouard 7842: } /* end simple */
7843: } /* end j */
1.223 brouard 7844: }else{
7845: i=i-ncovmodel;
7846: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7847: fprintf(ficgp," (1.");
7848: }
1.227 brouard 7849:
1.223 brouard 7850: if(ng != 1){
7851: fprintf(ficgp,")/(1");
1.227 brouard 7852:
1.264 brouard 7853: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7854: if(nagesqr==0)
1.264 brouard 7855: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7856: else /* nagesqr =1 */
1.264 brouard 7857: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1,k3+(cpt-1)*ncovmodel+1+nagesqr);
1.217 brouard 7858:
1.223 brouard 7859: ij=1;
7860: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7861: if(cptcovage >0){
7862: if((j-2)==Tage[ij]) { /* Bug valgrind */
7863: if(ij <=cptcovage) { /* Bug valgrind */
7864: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7865: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7866: ij++;
7867: }
7868: }
7869: }else
7870: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 7871: }
7872: fprintf(ficgp,")");
7873: }
7874: fprintf(ficgp,")");
7875: if(ng ==2)
7876: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7877: else /* ng= 3 */
7878: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7879: }else{ /* end ng <> 1 */
7880: if( k !=k2) /* logit p11 is hard to draw */
7881: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7882: }
7883: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7884: fprintf(ficgp,",");
7885: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7886: fprintf(ficgp,",");
7887: i=i+ncovmodel;
7888: } /* end k */
7889: } /* end k2 */
1.264 brouard 7890: fprintf(ficgp,"\n set out; unset label;\n");
7891: } /* end k1 */
1.223 brouard 7892: } /* end ng */
7893: /* avoid: */
7894: fflush(ficgp);
1.126 brouard 7895: } /* end gnuplot */
7896:
7897:
7898: /*************** Moving average **************/
1.219 brouard 7899: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7900: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7901:
1.222 brouard 7902: int i, cpt, cptcod;
7903: int modcovmax =1;
7904: int mobilavrange, mob;
7905: int iage=0;
7906:
1.266 brouard 7907: double sum=0., sumr=0.;
1.222 brouard 7908: double age;
1.266 brouard 7909: double *sumnewp, *sumnewm, *sumnewmr;
7910: double *agemingood, *agemaxgood;
7911: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7912:
7913:
1.225 brouard 7914: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7915: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7916:
7917: sumnewp = vector(1,ncovcombmax);
7918: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7919: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7920: agemingood = vector(1,ncovcombmax);
1.266 brouard 7921: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7922: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7923: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7924:
7925: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7926: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7927: sumnewp[cptcod]=0.;
1.266 brouard 7928: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7929: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7930: }
7931: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7932:
1.266 brouard 7933: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7934: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7935: else mobilavrange=mobilav;
7936: for (age=bage; age<=fage; age++)
7937: for (i=1; i<=nlstate;i++)
7938: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7939: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7940: /* We keep the original values on the extreme ages bage, fage and for
7941: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7942: we use a 5 terms etc. until the borders are no more concerned.
7943: */
7944: for (mob=3;mob <=mobilavrange;mob=mob+2){
7945: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7946: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7947: sumnewm[cptcod]=0.;
7948: for (i=1; i<=nlstate;i++){
1.222 brouard 7949: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7950: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7951: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7952: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7953: }
7954: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7955: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7956: } /* end i */
7957: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7958: } /* end cptcod */
1.222 brouard 7959: }/* end age */
7960: }/* end mob */
1.266 brouard 7961: }else{
7962: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7963: return -1;
1.266 brouard 7964: }
7965:
7966: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7967: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7968: if(invalidvarcomb[cptcod]){
7969: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7970: continue;
7971: }
1.219 brouard 7972:
1.266 brouard 7973: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
7974: sumnewm[cptcod]=0.;
7975: sumnewmr[cptcod]=0.;
7976: for (i=1; i<=nlstate;i++){
7977: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7978: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7979: }
7980: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7981: agemingoodr[cptcod]=age;
7982: }
7983: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7984: agemingood[cptcod]=age;
7985: }
7986: } /* age */
7987: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 7988: sumnewm[cptcod]=0.;
1.266 brouard 7989: sumnewmr[cptcod]=0.;
1.222 brouard 7990: for (i=1; i<=nlstate;i++){
7991: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 7992: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7993: }
7994: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7995: agemaxgoodr[cptcod]=age;
1.222 brouard 7996: }
7997: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 7998: agemaxgood[cptcod]=age;
7999: }
8000: } /* age */
8001: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8002: /* but they will change */
8003: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8004: sumnewm[cptcod]=0.;
8005: sumnewmr[cptcod]=0.;
8006: for (i=1; i<=nlstate;i++){
8007: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8008: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8009: }
8010: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8011: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8012: agemaxgoodr[cptcod]=age; /* age min */
8013: for (i=1; i<=nlstate;i++)
8014: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8015: }else{ /* bad we change the value with the values of good ages */
8016: for (i=1; i<=nlstate;i++){
8017: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8018: } /* i */
8019: } /* end bad */
8020: }else{
8021: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8022: agemaxgood[cptcod]=age;
8023: }else{ /* bad we change the value with the values of good ages */
8024: for (i=1; i<=nlstate;i++){
8025: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8026: } /* i */
8027: } /* end bad */
8028: }/* end else */
8029: sum=0.;sumr=0.;
8030: for (i=1; i<=nlstate;i++){
8031: sum+=mobaverage[(int)age][i][cptcod];
8032: sumr+=probs[(int)age][i][cptcod];
8033: }
8034: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8035: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 8036: } /* end bad */
8037: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8038: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8039: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 8040: } /* end bad */
8041: }/* age */
1.266 brouard 8042:
8043: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8044: sumnewm[cptcod]=0.;
1.266 brouard 8045: sumnewmr[cptcod]=0.;
1.222 brouard 8046: for (i=1; i<=nlstate;i++){
8047: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8048: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8049: }
8050: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8051: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8052: agemingoodr[cptcod]=age;
8053: for (i=1; i<=nlstate;i++)
8054: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8055: }else{ /* bad we change the value with the values of good ages */
8056: for (i=1; i<=nlstate;i++){
8057: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8058: } /* i */
8059: } /* end bad */
8060: }else{
8061: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8062: agemingood[cptcod]=age;
8063: }else{ /* bad */
8064: for (i=1; i<=nlstate;i++){
8065: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8066: } /* i */
8067: } /* end bad */
8068: }/* end else */
8069: sum=0.;sumr=0.;
8070: for (i=1; i<=nlstate;i++){
8071: sum+=mobaverage[(int)age][i][cptcod];
8072: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8073: }
1.266 brouard 8074: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8075: printf("Moving average B1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you decrease fage=%d?\n",cptcod, sum, (int) age, (int)fage);
1.266 brouard 8076: } /* end bad */
8077: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8078: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8079: printf("Moving average B2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase fage=%d\n",cptcod,sumr, (int)age, (int)fage);
1.222 brouard 8080: } /* end bad */
8081: }/* age */
1.266 brouard 8082:
1.222 brouard 8083:
8084: for (age=bage; age<=fage; age++){
1.235 brouard 8085: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8086: sumnewp[cptcod]=0.;
8087: sumnewm[cptcod]=0.;
8088: for (i=1; i<=nlstate;i++){
8089: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8090: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8091: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8092: }
8093: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8094: }
8095: /* printf("\n"); */
8096: /* } */
1.266 brouard 8097:
1.222 brouard 8098: /* brutal averaging */
1.266 brouard 8099: /* for (i=1; i<=nlstate;i++){ */
8100: /* for (age=1; age<=bage; age++){ */
8101: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8102: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8103: /* } */
8104: /* for (age=fage; age<=AGESUP; age++){ */
8105: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8106: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8107: /* } */
8108: /* } /\* end i status *\/ */
8109: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8110: /* for (age=1; age<=AGESUP; age++){ */
8111: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8112: /* mobaverage[(int)age][i][cptcod]=0.; */
8113: /* } */
8114: /* } */
1.222 brouard 8115: }/* end cptcod */
1.266 brouard 8116: free_vector(agemaxgoodr,1, ncovcombmax);
8117: free_vector(agemaxgood,1, ncovcombmax);
8118: free_vector(agemingood,1, ncovcombmax);
8119: free_vector(agemingoodr,1, ncovcombmax);
8120: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8121: free_vector(sumnewm,1, ncovcombmax);
8122: free_vector(sumnewp,1, ncovcombmax);
8123: return 0;
8124: }/* End movingaverage */
1.218 brouard 8125:
1.126 brouard 8126:
8127: /************** Forecasting ******************/
1.269 brouard 8128: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 8129: /* proj1, year, month, day of starting projection
8130: agemin, agemax range of age
8131: dateprev1 dateprev2 range of dates during which prevalence is computed
8132: anproj2 year of en of projection (same day and month as proj1).
8133: */
1.267 brouard 8134: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8135: double agec; /* generic age */
8136: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8137: double *popeffectif,*popcount;
8138: double ***p3mat;
1.218 brouard 8139: /* double ***mobaverage; */
1.126 brouard 8140: char fileresf[FILENAMELENGTH];
8141:
8142: agelim=AGESUP;
1.211 brouard 8143: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8144: in each health status at the date of interview (if between dateprev1 and dateprev2).
8145: We still use firstpass and lastpass as another selection.
8146: */
1.214 brouard 8147: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8148: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8149:
1.201 brouard 8150: strcpy(fileresf,"F_");
8151: strcat(fileresf,fileresu);
1.126 brouard 8152: if((ficresf=fopen(fileresf,"w"))==NULL) {
8153: printf("Problem with forecast resultfile: %s\n", fileresf);
8154: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8155: }
1.235 brouard 8156: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8157: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8158:
1.225 brouard 8159: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8160:
8161:
8162: stepsize=(int) (stepm+YEARM-1)/YEARM;
8163: if (stepm<=12) stepsize=1;
8164: if(estepm < stepm){
8165: printf ("Problem %d lower than %d\n",estepm, stepm);
8166: }
1.270 brouard 8167: else{
8168: hstepm=estepm;
8169: }
8170: if(estepm > stepm){ /* Yes every two year */
8171: stepsize=2;
8172: }
1.126 brouard 8173:
8174: hstepm=hstepm/stepm;
8175: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8176: fractional in yp1 */
8177: anprojmean=yp;
8178: yp2=modf((yp1*12),&yp);
8179: mprojmean=yp;
8180: yp1=modf((yp2*30.5),&yp);
8181: jprojmean=yp;
8182: if(jprojmean==0) jprojmean=1;
8183: if(mprojmean==0) jprojmean=1;
8184:
1.227 brouard 8185: i1=pow(2,cptcoveff);
1.126 brouard 8186: if (cptcovn < 1){i1=1;}
8187:
8188: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8189:
8190: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8191:
1.126 brouard 8192: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8193: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8194: for(k=1; k<=i1;k++){
1.253 brouard 8195: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8196: continue;
1.227 brouard 8197: if(invalidvarcomb[k]){
8198: printf("\nCombination (%d) projection ignored because no cases \n",k);
8199: continue;
8200: }
8201: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8202: for(j=1;j<=cptcoveff;j++) {
8203: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8204: }
1.235 brouard 8205: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8206: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8207: }
1.227 brouard 8208: fprintf(ficresf," yearproj age");
8209: for(j=1; j<=nlstate+ndeath;j++){
8210: for(i=1; i<=nlstate;i++)
8211: fprintf(ficresf," p%d%d",i,j);
8212: fprintf(ficresf," wp.%d",j);
8213: }
8214: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8215: fprintf(ficresf,"\n");
8216: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8217: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8218: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8219: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8220: nhstepm = nhstepm/hstepm;
8221: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8222: oldm=oldms;savm=savms;
1.268 brouard 8223: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8224: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8225: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8226: for (h=0; h<=nhstepm; h++){
8227: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8228: break;
8229: }
8230: }
8231: fprintf(ficresf,"\n");
8232: for(j=1;j<=cptcoveff;j++)
8233: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8234: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8235:
8236: for(j=1; j<=nlstate+ndeath;j++) {
8237: ppij=0.;
8238: for(i=1; i<=nlstate;i++) {
8239: /* if (mobilav>=1) */
1.269 brouard 8240: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
1.268 brouard 8241: /* else { */ /* even if mobilav==-1 we use mobaverage */
8242: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8243: /* } */
8244: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8245: } /* end i */
8246: fprintf(ficresf," %.3f", ppij);
8247: }/* end j */
1.227 brouard 8248: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8249: } /* end agec */
1.266 brouard 8250: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8251: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8252: } /* end yearp */
8253: } /* end k */
1.219 brouard 8254:
1.126 brouard 8255: fclose(ficresf);
1.215 brouard 8256: printf("End of Computing forecasting \n");
8257: fprintf(ficlog,"End of Computing forecasting\n");
8258:
1.126 brouard 8259: }
8260:
1.269 brouard 8261: /************** Back Forecasting ******************/
8262: void prevbackforecast(char fileres[], double ***prevacurrent, double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){
1.267 brouard 8263: /* back1, year, month, day of starting backection
8264: agemin, agemax range of age
8265: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8266: anback2 year of end of backprojection (same day and month as back1).
8267: prevacurrent and prev are prevalences.
1.267 brouard 8268: */
8269: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8270: double agec; /* generic age */
1.268 brouard 8271: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8272: double *popeffectif,*popcount;
8273: double ***p3mat;
8274: /* double ***mobaverage; */
8275: char fileresfb[FILENAMELENGTH];
8276:
1.268 brouard 8277: agelim=AGEINF;
1.267 brouard 8278: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8279: in each health status at the date of interview (if between dateprev1 and dateprev2).
8280: We still use firstpass and lastpass as another selection.
8281: */
8282: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8283: /* firstpass, lastpass, stepm, weightopt, model); */
8284:
8285: /*Do we need to compute prevalence again?*/
8286:
8287: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8288:
8289: strcpy(fileresfb,"FB_");
8290: strcat(fileresfb,fileresu);
8291: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8292: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8293: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8294: }
8295: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8296: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8297:
8298: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8299:
8300:
8301: stepsize=(int) (stepm+YEARM-1)/YEARM;
8302: if (stepm<=12) stepsize=1;
8303: if(estepm < stepm){
8304: printf ("Problem %d lower than %d\n",estepm, stepm);
8305: }
1.270 brouard 8306: else{
8307: hstepm=estepm;
8308: }
8309: if(estepm >= stepm){ /* Yes every two year */
8310: stepsize=2;
8311: }
1.267 brouard 8312:
8313: hstepm=hstepm/stepm;
8314: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8315: fractional in yp1 */
8316: anprojmean=yp;
8317: yp2=modf((yp1*12),&yp);
8318: mprojmean=yp;
8319: yp1=modf((yp2*30.5),&yp);
8320: jprojmean=yp;
8321: if(jprojmean==0) jprojmean=1;
8322: if(mprojmean==0) jprojmean=1;
8323:
8324: i1=pow(2,cptcoveff);
8325: if (cptcovn < 1){i1=1;}
8326:
8327: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8328: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8329:
8330: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8331:
8332: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8333: for(k=1; k<=i1;k++){
8334: if(i1 != 1 && TKresult[nres]!= k)
8335: continue;
8336: if(invalidvarcomb[k]){
8337: printf("\nCombination (%d) projection ignored because no cases \n",k);
8338: continue;
8339: }
1.268 brouard 8340: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8341: for(j=1;j<=cptcoveff;j++) {
8342: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8343: }
8344: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8345: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8346: }
8347: fprintf(ficresfb," yearbproj age");
8348: for(j=1; j<=nlstate+ndeath;j++){
8349: for(i=1; i<=nlstate;i++)
1.268 brouard 8350: fprintf(ficresfb," b%d%d",i,j);
8351: fprintf(ficresfb," b.%d",j);
1.267 brouard 8352: }
8353: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8354: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8355: fprintf(ficresfb,"\n");
8356: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 ! brouard 8357: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8358: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8359: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8360: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8361: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8362: nhstepm = nhstepm/hstepm;
8363: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8364: oldm=oldms;savm=savms;
1.268 brouard 8365: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8366: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8367: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8368: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8369: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8370: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8371: for (h=0; h<=nhstepm; h++){
1.268 brouard 8372: if (h*hstepm/YEARM*stepm ==-yearp) {
8373: break;
8374: }
8375: }
8376: fprintf(ficresfb,"\n");
8377: for(j=1;j<=cptcoveff;j++)
8378: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8379: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8380: for(i=1; i<=nlstate+ndeath;i++) {
8381: ppij=0.;ppi=0.;
8382: for(j=1; j<=nlstate;j++) {
8383: /* if (mobilav==1) */
1.269 brouard 8384: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8385: ppi=ppi+prevacurrent[(int)agec][j][k];
8386: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8387: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8388: /* else { */
8389: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8390: /* } */
1.268 brouard 8391: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8392: } /* end j */
8393: if(ppi <0.99){
8394: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8395: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8396: }
8397: fprintf(ficresfb," %.3f", ppij);
8398: }/* end j */
1.267 brouard 8399: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8400: } /* end agec */
8401: } /* end yearp */
8402: } /* end k */
1.217 brouard 8403:
1.267 brouard 8404: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8405:
1.267 brouard 8406: fclose(ficresfb);
8407: printf("End of Computing Back forecasting \n");
8408: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8409:
1.267 brouard 8410: }
1.217 brouard 8411:
1.269 brouard 8412: /* Variance of prevalence limit: varprlim */
8413: void varprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **prlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
8414: /*------- Variance of period (stable) prevalence------*/
8415:
8416: char fileresvpl[FILENAMELENGTH];
8417: FILE *ficresvpl;
8418: double **oldm, **savm;
8419: double **varpl; /* Variances of prevalence limits by age */
8420: int i1, k, nres, j ;
8421:
8422: strcpy(fileresvpl,"VPL_");
8423: strcat(fileresvpl,fileresu);
8424: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8425: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8426: exit(0);
8427: }
8428: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8429: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8430:
8431: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8432: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8433:
8434: i1=pow(2,cptcoveff);
8435: if (cptcovn < 1){i1=1;}
8436:
8437: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8438: for(k=1; k<=i1;k++){
8439: if(i1 != 1 && TKresult[nres]!= k)
8440: continue;
8441: fprintf(ficresvpl,"\n#****** ");
8442: printf("\n#****** ");
8443: fprintf(ficlog,"\n#****** ");
8444: for(j=1;j<=cptcoveff;j++) {
8445: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8446: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8447: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8448: }
8449: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8450: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8451: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8452: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8453: }
8454: fprintf(ficresvpl,"******\n");
8455: printf("******\n");
8456: fprintf(ficlog,"******\n");
8457:
8458: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8459: oldm=oldms;savm=savms;
8460: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8461: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8462: /*}*/
8463: }
8464:
8465: fclose(ficresvpl);
8466: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8467: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8468:
8469: }
8470: /* Variance of back prevalence: varbprlim */
8471: void varbprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **bprlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
8472: /*------- Variance of back (stable) prevalence------*/
8473:
8474: char fileresvbl[FILENAMELENGTH];
8475: FILE *ficresvbl;
8476:
8477: double **oldm, **savm;
8478: double **varbpl; /* Variances of back prevalence limits by age */
8479: int i1, k, nres, j ;
8480:
8481: strcpy(fileresvbl,"VBL_");
8482: strcat(fileresvbl,fileresu);
8483: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8484: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8485: exit(0);
8486: }
8487: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8488: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8489:
8490:
8491: i1=pow(2,cptcoveff);
8492: if (cptcovn < 1){i1=1;}
8493:
8494: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8495: for(k=1; k<=i1;k++){
8496: if(i1 != 1 && TKresult[nres]!= k)
8497: continue;
8498: fprintf(ficresvbl,"\n#****** ");
8499: printf("\n#****** ");
8500: fprintf(ficlog,"\n#****** ");
8501: for(j=1;j<=cptcoveff;j++) {
8502: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8503: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8504: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8505: }
8506: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8507: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8508: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8509: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8510: }
8511: fprintf(ficresvbl,"******\n");
8512: printf("******\n");
8513: fprintf(ficlog,"******\n");
8514:
8515: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8516: oldm=oldms;savm=savms;
8517:
8518: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8519: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8520: /*}*/
8521: }
8522:
8523: fclose(ficresvbl);
8524: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8525: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8526:
8527: } /* End of varbprlim */
8528:
1.126 brouard 8529: /************** Forecasting *****not tested NB*************/
1.227 brouard 8530: /* 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 8531:
1.227 brouard 8532: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8533: /* int *popage; */
8534: /* double calagedatem, agelim, kk1, kk2; */
8535: /* double *popeffectif,*popcount; */
8536: /* double ***p3mat,***tabpop,***tabpopprev; */
8537: /* /\* double ***mobaverage; *\/ */
8538: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8539:
1.227 brouard 8540: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8541: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8542: /* agelim=AGESUP; */
8543: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8544:
1.227 brouard 8545: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8546:
8547:
1.227 brouard 8548: /* strcpy(filerespop,"POP_"); */
8549: /* strcat(filerespop,fileresu); */
8550: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8551: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8552: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8553: /* } */
8554: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8555: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8556:
1.227 brouard 8557: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8558:
1.227 brouard 8559: /* /\* if (mobilav!=0) { *\/ */
8560: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8561: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8562: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8563: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8564: /* /\* } *\/ */
8565: /* /\* } *\/ */
1.126 brouard 8566:
1.227 brouard 8567: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8568: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8569:
1.227 brouard 8570: /* agelim=AGESUP; */
1.126 brouard 8571:
1.227 brouard 8572: /* hstepm=1; */
8573: /* hstepm=hstepm/stepm; */
1.218 brouard 8574:
1.227 brouard 8575: /* if (popforecast==1) { */
8576: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8577: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8578: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8579: /* } */
8580: /* popage=ivector(0,AGESUP); */
8581: /* popeffectif=vector(0,AGESUP); */
8582: /* popcount=vector(0,AGESUP); */
1.126 brouard 8583:
1.227 brouard 8584: /* i=1; */
8585: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8586:
1.227 brouard 8587: /* imx=i; */
8588: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8589: /* } */
1.218 brouard 8590:
1.227 brouard 8591: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8592: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8593: /* k=k+1; */
8594: /* fprintf(ficrespop,"\n#******"); */
8595: /* for(j=1;j<=cptcoveff;j++) { */
8596: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8597: /* } */
8598: /* fprintf(ficrespop,"******\n"); */
8599: /* fprintf(ficrespop,"# Age"); */
8600: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8601: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8602:
1.227 brouard 8603: /* for (cpt=0; cpt<=0;cpt++) { */
8604: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8605:
1.227 brouard 8606: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8607: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8608: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8609:
1.227 brouard 8610: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8611: /* oldm=oldms;savm=savms; */
8612: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8613:
1.227 brouard 8614: /* for (h=0; h<=nhstepm; h++){ */
8615: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8616: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8617: /* } */
8618: /* for(j=1; j<=nlstate+ndeath;j++) { */
8619: /* kk1=0.;kk2=0; */
8620: /* for(i=1; i<=nlstate;i++) { */
8621: /* if (mobilav==1) */
8622: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8623: /* else { */
8624: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8625: /* } */
8626: /* } */
8627: /* if (h==(int)(calagedatem+12*cpt)){ */
8628: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8629: /* /\*fprintf(ficrespop," %.3f", kk1); */
8630: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8631: /* } */
8632: /* } */
8633: /* for(i=1; i<=nlstate;i++){ */
8634: /* kk1=0.; */
8635: /* for(j=1; j<=nlstate;j++){ */
8636: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8637: /* } */
8638: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8639: /* } */
1.218 brouard 8640:
1.227 brouard 8641: /* if (h==(int)(calagedatem+12*cpt)) */
8642: /* for(j=1; j<=nlstate;j++) */
8643: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8644: /* } */
8645: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8646: /* } */
8647: /* } */
1.218 brouard 8648:
1.227 brouard 8649: /* /\******\/ */
1.218 brouard 8650:
1.227 brouard 8651: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8652: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8653: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8654: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8655: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8656:
1.227 brouard 8657: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8658: /* oldm=oldms;savm=savms; */
8659: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8660: /* for (h=0; h<=nhstepm; h++){ */
8661: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8662: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8663: /* } */
8664: /* for(j=1; j<=nlstate+ndeath;j++) { */
8665: /* kk1=0.;kk2=0; */
8666: /* for(i=1; i<=nlstate;i++) { */
8667: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8668: /* } */
8669: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8670: /* } */
8671: /* } */
8672: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8673: /* } */
8674: /* } */
8675: /* } */
8676: /* } */
1.218 brouard 8677:
1.227 brouard 8678: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8679:
1.227 brouard 8680: /* if (popforecast==1) { */
8681: /* free_ivector(popage,0,AGESUP); */
8682: /* free_vector(popeffectif,0,AGESUP); */
8683: /* free_vector(popcount,0,AGESUP); */
8684: /* } */
8685: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8686: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8687: /* fclose(ficrespop); */
8688: /* } /\* End of popforecast *\/ */
1.218 brouard 8689:
1.126 brouard 8690: int fileappend(FILE *fichier, char *optionfich)
8691: {
8692: if((fichier=fopen(optionfich,"a"))==NULL) {
8693: printf("Problem with file: %s\n", optionfich);
8694: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8695: return (0);
8696: }
8697: fflush(fichier);
8698: return (1);
8699: }
8700:
8701:
8702: /**************** function prwizard **********************/
8703: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8704: {
8705:
8706: /* Wizard to print covariance matrix template */
8707:
1.164 brouard 8708: char ca[32], cb[32];
8709: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8710: int numlinepar;
8711:
8712: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8713: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8714: for(i=1; i <=nlstate; i++){
8715: jj=0;
8716: for(j=1; j <=nlstate+ndeath; j++){
8717: if(j==i) continue;
8718: jj++;
8719: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8720: printf("%1d%1d",i,j);
8721: fprintf(ficparo,"%1d%1d",i,j);
8722: for(k=1; k<=ncovmodel;k++){
8723: /* printf(" %lf",param[i][j][k]); */
8724: /* fprintf(ficparo," %lf",param[i][j][k]); */
8725: printf(" 0.");
8726: fprintf(ficparo," 0.");
8727: }
8728: printf("\n");
8729: fprintf(ficparo,"\n");
8730: }
8731: }
8732: printf("# Scales (for hessian or gradient estimation)\n");
8733: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8734: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8735: for(i=1; i <=nlstate; i++){
8736: jj=0;
8737: for(j=1; j <=nlstate+ndeath; j++){
8738: if(j==i) continue;
8739: jj++;
8740: fprintf(ficparo,"%1d%1d",i,j);
8741: printf("%1d%1d",i,j);
8742: fflush(stdout);
8743: for(k=1; k<=ncovmodel;k++){
8744: /* printf(" %le",delti3[i][j][k]); */
8745: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8746: printf(" 0.");
8747: fprintf(ficparo," 0.");
8748: }
8749: numlinepar++;
8750: printf("\n");
8751: fprintf(ficparo,"\n");
8752: }
8753: }
8754: printf("# Covariance matrix\n");
8755: /* # 121 Var(a12)\n\ */
8756: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8757: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8758: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8759: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8760: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8761: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8762: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8763: fflush(stdout);
8764: fprintf(ficparo,"# Covariance matrix\n");
8765: /* # 121 Var(a12)\n\ */
8766: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8767: /* # ...\n\ */
8768: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8769:
8770: for(itimes=1;itimes<=2;itimes++){
8771: jj=0;
8772: for(i=1; i <=nlstate; i++){
8773: for(j=1; j <=nlstate+ndeath; j++){
8774: if(j==i) continue;
8775: for(k=1; k<=ncovmodel;k++){
8776: jj++;
8777: ca[0]= k+'a'-1;ca[1]='\0';
8778: if(itimes==1){
8779: printf("#%1d%1d%d",i,j,k);
8780: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8781: }else{
8782: printf("%1d%1d%d",i,j,k);
8783: fprintf(ficparo,"%1d%1d%d",i,j,k);
8784: /* printf(" %.5le",matcov[i][j]); */
8785: }
8786: ll=0;
8787: for(li=1;li <=nlstate; li++){
8788: for(lj=1;lj <=nlstate+ndeath; lj++){
8789: if(lj==li) continue;
8790: for(lk=1;lk<=ncovmodel;lk++){
8791: ll++;
8792: if(ll<=jj){
8793: cb[0]= lk +'a'-1;cb[1]='\0';
8794: if(ll<jj){
8795: if(itimes==1){
8796: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8797: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8798: }else{
8799: printf(" 0.");
8800: fprintf(ficparo," 0.");
8801: }
8802: }else{
8803: if(itimes==1){
8804: printf(" Var(%s%1d%1d)",ca,i,j);
8805: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8806: }else{
8807: printf(" 0.");
8808: fprintf(ficparo," 0.");
8809: }
8810: }
8811: }
8812: } /* end lk */
8813: } /* end lj */
8814: } /* end li */
8815: printf("\n");
8816: fprintf(ficparo,"\n");
8817: numlinepar++;
8818: } /* end k*/
8819: } /*end j */
8820: } /* end i */
8821: } /* end itimes */
8822:
8823: } /* end of prwizard */
8824: /******************* Gompertz Likelihood ******************************/
8825: double gompertz(double x[])
8826: {
8827: double A,B,L=0.0,sump=0.,num=0.;
8828: int i,n=0; /* n is the size of the sample */
8829:
1.220 brouard 8830: for (i=1;i<=imx ; i++) {
1.126 brouard 8831: sump=sump+weight[i];
8832: /* sump=sump+1;*/
8833: num=num+1;
8834: }
8835:
8836:
8837: /* for (i=0; i<=imx; i++)
8838: 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]);*/
8839:
8840: for (i=1;i<=imx ; i++)
8841: {
8842: if (cens[i] == 1 && wav[i]>1)
8843: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8844:
8845: if (cens[i] == 0 && wav[i]>1)
8846: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8847: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8848:
8849: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8850: if (wav[i] > 1 ) { /* ??? */
8851: L=L+A*weight[i];
8852: /* 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]);*/
8853: }
8854: }
8855:
8856: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8857:
8858: return -2*L*num/sump;
8859: }
8860:
1.136 brouard 8861: #ifdef GSL
8862: /******************* Gompertz_f Likelihood ******************************/
8863: double gompertz_f(const gsl_vector *v, void *params)
8864: {
8865: double A,B,LL=0.0,sump=0.,num=0.;
8866: double *x= (double *) v->data;
8867: int i,n=0; /* n is the size of the sample */
8868:
8869: for (i=0;i<=imx-1 ; i++) {
8870: sump=sump+weight[i];
8871: /* sump=sump+1;*/
8872: num=num+1;
8873: }
8874:
8875:
8876: /* for (i=0; i<=imx; i++)
8877: 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]);*/
8878: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8879: for (i=1;i<=imx ; i++)
8880: {
8881: if (cens[i] == 1 && wav[i]>1)
8882: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8883:
8884: if (cens[i] == 0 && wav[i]>1)
8885: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8886: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8887:
8888: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8889: if (wav[i] > 1 ) { /* ??? */
8890: LL=LL+A*weight[i];
8891: /* 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]);*/
8892: }
8893: }
8894:
8895: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8896: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8897:
8898: return -2*LL*num/sump;
8899: }
8900: #endif
8901:
1.126 brouard 8902: /******************* Printing html file ***********/
1.201 brouard 8903: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8904: int lastpass, int stepm, int weightopt, char model[],\
8905: int imx, double p[],double **matcov,double agemortsup){
8906: int i,k;
8907:
8908: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8909: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8910: for (i=1;i<=2;i++)
8911: 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 8912: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8913: fprintf(fichtm,"</ul>");
8914:
8915: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8916:
8917: 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>");
8918:
8919: for (k=agegomp;k<(agemortsup-2);k++)
8920: 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]);
8921:
8922:
8923: fflush(fichtm);
8924: }
8925:
8926: /******************* Gnuplot file **************/
1.201 brouard 8927: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8928:
8929: char dirfileres[132],optfileres[132];
1.164 brouard 8930:
1.126 brouard 8931: int ng;
8932:
8933:
8934: /*#ifdef windows */
8935: fprintf(ficgp,"cd \"%s\" \n",pathc);
8936: /*#endif */
8937:
8938:
8939: strcpy(dirfileres,optionfilefiname);
8940: strcpy(optfileres,"vpl");
1.199 brouard 8941: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8942: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8943: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8944: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8945: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8946:
8947: }
8948:
1.136 brouard 8949: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8950: {
1.126 brouard 8951:
1.136 brouard 8952: /*-------- data file ----------*/
8953: FILE *fic;
8954: char dummy[]=" ";
1.240 brouard 8955: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8956: int lstra;
1.136 brouard 8957: int linei, month, year,iout;
8958: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8959: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8960: char *stratrunc;
1.223 brouard 8961:
1.240 brouard 8962: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8963: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8964:
1.240 brouard 8965: for(v=1; v <=ncovcol;v++){
8966: DummyV[v]=0;
8967: FixedV[v]=0;
8968: }
8969: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8970: DummyV[v]=1;
8971: FixedV[v]=0;
8972: }
8973: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8974: DummyV[v]=0;
8975: FixedV[v]=1;
8976: }
8977: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8978: DummyV[v]=1;
8979: FixedV[v]=1;
8980: }
8981: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8982: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8983: 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]);
8984: }
1.126 brouard 8985:
1.136 brouard 8986: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8987: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8988: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8989: }
1.126 brouard 8990:
1.136 brouard 8991: i=1;
8992: linei=0;
8993: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8994: linei=linei+1;
8995: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8996: if(line[j] == '\t')
8997: line[j] = ' ';
8998: }
8999: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9000: ;
9001: };
9002: line[j+1]=0; /* Trims blanks at end of line */
9003: if(line[0]=='#'){
9004: fprintf(ficlog,"Comment line\n%s\n",line);
9005: printf("Comment line\n%s\n",line);
9006: continue;
9007: }
9008: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9009: strcpy(line, linetmp);
1.223 brouard 9010:
9011: /* Loops on waves */
9012: for (j=maxwav;j>=1;j--){
9013: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9014: cutv(stra, strb, line, ' ');
9015: if(strb[0]=='.') { /* Missing value */
9016: lval=-1;
9017: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9018: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9019: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9020: 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);
9021: 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);
9022: return 1;
9023: }
9024: }else{
9025: errno=0;
9026: /* what_kind_of_number(strb); */
9027: dval=strtod(strb,&endptr);
9028: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9029: /* if(strb != endptr && *endptr == '\0') */
9030: /* dval=dlval; */
9031: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9032: if( strb[0]=='\0' || (*endptr != '\0')){
9033: 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);
9034: 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);
9035: return 1;
9036: }
9037: cotqvar[j][iv][i]=dval;
9038: cotvar[j][ntv+iv][i]=dval;
9039: }
9040: strcpy(line,stra);
1.223 brouard 9041: }/* end loop ntqv */
1.225 brouard 9042:
1.223 brouard 9043: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9044: cutv(stra, strb, line, ' ');
9045: if(strb[0]=='.') { /* Missing value */
9046: lval=-1;
9047: }else{
9048: errno=0;
9049: lval=strtol(strb,&endptr,10);
9050: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9051: if( strb[0]=='\0' || (*endptr != '\0')){
9052: 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);
9053: 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);
9054: return 1;
9055: }
9056: }
9057: if(lval <-1 || lval >1){
9058: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9059: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9060: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9061: For example, for multinomial values like 1, 2 and 3,\n \
9062: build V1=0 V2=0 for the reference value (1),\n \
9063: V1=1 V2=0 for (2) \n \
1.223 brouard 9064: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9065: output of IMaCh is often meaningless.\n \
1.223 brouard 9066: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9067: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9068: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9069: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9070: For example, for multinomial values like 1, 2 and 3,\n \
9071: build V1=0 V2=0 for the reference value (1),\n \
9072: V1=1 V2=0 for (2) \n \
1.223 brouard 9073: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9074: output of IMaCh is often meaningless.\n \
1.223 brouard 9075: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9076: return 1;
9077: }
9078: cotvar[j][iv][i]=(double)(lval);
9079: strcpy(line,stra);
1.223 brouard 9080: }/* end loop ntv */
1.225 brouard 9081:
1.223 brouard 9082: /* Statuses at wave */
1.137 brouard 9083: cutv(stra, strb, line, ' ');
1.223 brouard 9084: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9085: lval=-1;
1.136 brouard 9086: }else{
1.238 brouard 9087: errno=0;
9088: lval=strtol(strb,&endptr,10);
9089: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9090: if( strb[0]=='\0' || (*endptr != '\0')){
9091: 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);
9092: 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);
9093: return 1;
9094: }
1.136 brouard 9095: }
1.225 brouard 9096:
1.136 brouard 9097: s[j][i]=lval;
1.225 brouard 9098:
1.223 brouard 9099: /* Date of Interview */
1.136 brouard 9100: strcpy(line,stra);
9101: cutv(stra, strb,line,' ');
1.169 brouard 9102: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9103: }
1.169 brouard 9104: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9105: month=99;
9106: year=9999;
1.136 brouard 9107: }else{
1.225 brouard 9108: 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);
9109: 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);
9110: return 1;
1.136 brouard 9111: }
9112: anint[j][i]= (double) year;
9113: mint[j][i]= (double)month;
9114: strcpy(line,stra);
1.223 brouard 9115: } /* End loop on waves */
1.225 brouard 9116:
1.223 brouard 9117: /* Date of death */
1.136 brouard 9118: cutv(stra, strb,line,' ');
1.169 brouard 9119: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9120: }
1.169 brouard 9121: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9122: month=99;
9123: year=9999;
9124: }else{
1.141 brouard 9125: 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 9126: 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);
9127: return 1;
1.136 brouard 9128: }
9129: andc[i]=(double) year;
9130: moisdc[i]=(double) month;
9131: strcpy(line,stra);
9132:
1.223 brouard 9133: /* Date of birth */
1.136 brouard 9134: cutv(stra, strb,line,' ');
1.169 brouard 9135: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9136: }
1.169 brouard 9137: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9138: month=99;
9139: year=9999;
9140: }else{
1.141 brouard 9141: 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);
9142: 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 9143: return 1;
1.136 brouard 9144: }
9145: if (year==9999) {
1.141 brouard 9146: 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);
9147: 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 9148: return 1;
9149:
1.136 brouard 9150: }
9151: annais[i]=(double)(year);
9152: moisnais[i]=(double)(month);
9153: strcpy(line,stra);
1.225 brouard 9154:
1.223 brouard 9155: /* Sample weight */
1.136 brouard 9156: cutv(stra, strb,line,' ');
9157: errno=0;
9158: dval=strtod(strb,&endptr);
9159: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9160: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9161: 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 9162: fflush(ficlog);
9163: return 1;
9164: }
9165: weight[i]=dval;
9166: strcpy(line,stra);
1.225 brouard 9167:
1.223 brouard 9168: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9169: cutv(stra, strb, line, ' ');
9170: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9171: lval=-1;
1.223 brouard 9172: }else{
1.225 brouard 9173: errno=0;
9174: /* what_kind_of_number(strb); */
9175: dval=strtod(strb,&endptr);
9176: /* if(strb != endptr && *endptr == '\0') */
9177: /* dval=dlval; */
9178: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9179: if( strb[0]=='\0' || (*endptr != '\0')){
9180: 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);
9181: 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);
9182: return 1;
9183: }
9184: coqvar[iv][i]=dval;
1.226 brouard 9185: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9186: }
9187: strcpy(line,stra);
9188: }/* end loop nqv */
1.136 brouard 9189:
1.223 brouard 9190: /* Covariate values */
1.136 brouard 9191: for (j=ncovcol;j>=1;j--){
9192: cutv(stra, strb,line,' ');
1.223 brouard 9193: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9194: lval=-1;
1.136 brouard 9195: }else{
1.225 brouard 9196: errno=0;
9197: lval=strtol(strb,&endptr,10);
9198: if( strb[0]=='\0' || (*endptr != '\0')){
9199: 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);
9200: 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);
9201: return 1;
9202: }
1.136 brouard 9203: }
9204: if(lval <-1 || lval >1){
1.225 brouard 9205: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9206: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9207: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9208: For example, for multinomial values like 1, 2 and 3,\n \
9209: build V1=0 V2=0 for the reference value (1),\n \
9210: V1=1 V2=0 for (2) \n \
1.136 brouard 9211: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9212: output of IMaCh is often meaningless.\n \
1.136 brouard 9213: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9214: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9215: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9216: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9217: For example, for multinomial values like 1, 2 and 3,\n \
9218: build V1=0 V2=0 for the reference value (1),\n \
9219: V1=1 V2=0 for (2) \n \
1.136 brouard 9220: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9221: output of IMaCh is often meaningless.\n \
1.136 brouard 9222: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9223: return 1;
1.136 brouard 9224: }
9225: covar[j][i]=(double)(lval);
9226: strcpy(line,stra);
9227: }
9228: lstra=strlen(stra);
1.225 brouard 9229:
1.136 brouard 9230: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9231: stratrunc = &(stra[lstra-9]);
9232: num[i]=atol(stratrunc);
9233: }
9234: else
9235: num[i]=atol(stra);
9236: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9237: 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;}*/
9238:
9239: i=i+1;
9240: } /* End loop reading data */
1.225 brouard 9241:
1.136 brouard 9242: *imax=i-1; /* Number of individuals */
9243: fclose(fic);
1.225 brouard 9244:
1.136 brouard 9245: return (0);
1.164 brouard 9246: /* endread: */
1.225 brouard 9247: printf("Exiting readdata: ");
9248: fclose(fic);
9249: return (1);
1.223 brouard 9250: }
1.126 brouard 9251:
1.234 brouard 9252: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9253: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9254: while (*p2 == ' ')
1.234 brouard 9255: p2++;
9256: /* while ((*p1++ = *p2++) !=0) */
9257: /* ; */
9258: /* do */
9259: /* while (*p2 == ' ') */
9260: /* p2++; */
9261: /* while (*p1++ == *p2++); */
9262: *stri=p2;
1.145 brouard 9263: }
9264:
1.235 brouard 9265: int decoderesult ( char resultline[], int nres)
1.230 brouard 9266: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9267: {
1.235 brouard 9268: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9269: char resultsav[MAXLINE];
1.234 brouard 9270: int resultmodel[MAXLINE];
9271: int modelresult[MAXLINE];
1.230 brouard 9272: char stra[80], strb[80], strc[80], strd[80],stre[80];
9273:
1.234 brouard 9274: removefirstspace(&resultline);
1.233 brouard 9275: printf("decoderesult:%s\n",resultline);
1.230 brouard 9276:
9277: if (strstr(resultline,"v") !=0){
9278: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9279: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9280: return 1;
9281: }
9282: trimbb(resultsav, resultline);
9283: if (strlen(resultsav) >1){
9284: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9285: }
1.253 brouard 9286: if(j == 0){ /* Resultline but no = */
9287: TKresult[nres]=0; /* Combination for the nresult and the model */
9288: return (0);
9289: }
9290:
1.234 brouard 9291: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9292: 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);
9293: 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);
9294: }
9295: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9296: if(nbocc(resultsav,'=') >1){
9297: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9298: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9299: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9300: }else
9301: cutl(strc,strd,resultsav,'=');
1.230 brouard 9302: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9303:
1.230 brouard 9304: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9305: Tvarsel[k]=atoi(strc);
9306: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9307: /* cptcovsel++; */
9308: if (nbocc(stra,'=') >0)
9309: strcpy(resultsav,stra); /* and analyzes it */
9310: }
1.235 brouard 9311: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9312: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9313: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9314: match=0;
1.236 brouard 9315: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9316: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9317: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9318: match=1;
9319: break;
9320: }
9321: }
9322: if(match == 0){
9323: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9324: }
9325: }
9326: }
1.235 brouard 9327: /* Checking for missing or useless values in comparison of current model needs */
9328: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9329: match=0;
1.235 brouard 9330: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9331: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9332: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9333: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9334: ++match;
9335: }
9336: }
9337: }
9338: if(match == 0){
9339: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9340: }else if(match > 1){
9341: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9342: }
9343: }
1.235 brouard 9344:
1.234 brouard 9345: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9346: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9347: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9348: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9349: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9350: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9351: /* 1 0 0 0 */
9352: /* 2 1 0 0 */
9353: /* 3 0 1 0 */
9354: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9355: /* 5 0 0 1 */
9356: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9357: /* 7 0 1 1 */
9358: /* 8 1 1 1 */
1.237 brouard 9359: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9360: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9361: /* V5*age V5 known which value for nres? */
9362: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9363: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9364: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9365: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9366: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9367: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9368: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9369: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9370: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9371: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9372: k4++;;
9373: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9374: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9375: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9376: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9377: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9378: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9379: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9380: k4q++;;
9381: }
9382: }
1.234 brouard 9383:
1.235 brouard 9384: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9385: return (0);
9386: }
1.235 brouard 9387:
1.230 brouard 9388: int decodemodel( char model[], int lastobs)
9389: /**< This routine decodes the model and returns:
1.224 brouard 9390: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9391: * - nagesqr = 1 if age*age in the model, otherwise 0.
9392: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9393: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9394: * - cptcovage number of covariates with age*products =2
9395: * - cptcovs number of simple covariates
9396: * - 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
9397: * which is a new column after the 9 (ncovcol) variables.
9398: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9399: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9400: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9401: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9402: */
1.136 brouard 9403: {
1.238 brouard 9404: int i, j, k, ks, v;
1.227 brouard 9405: int j1, k1, k2, k3, k4;
1.136 brouard 9406: char modelsav[80];
1.145 brouard 9407: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9408: char *strpt;
1.136 brouard 9409:
1.145 brouard 9410: /*removespace(model);*/
1.136 brouard 9411: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9412: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9413: if (strstr(model,"AGE") !=0){
1.192 brouard 9414: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9415: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9416: return 1;
9417: }
1.141 brouard 9418: if (strstr(model,"v") !=0){
9419: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9420: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9421: return 1;
9422: }
1.187 brouard 9423: strcpy(modelsav,model);
9424: if ((strpt=strstr(model,"age*age")) !=0){
9425: printf(" strpt=%s, model=%s\n",strpt, model);
9426: if(strpt != model){
1.234 brouard 9427: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9428: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9429: corresponding column of parameters.\n",model);
1.234 brouard 9430: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9431: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9432: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9433: return 1;
1.225 brouard 9434: }
1.187 brouard 9435: nagesqr=1;
9436: if (strstr(model,"+age*age") !=0)
1.234 brouard 9437: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9438: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9439: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9440: else
1.234 brouard 9441: substrchaine(modelsav, model, "age*age");
1.187 brouard 9442: }else
9443: nagesqr=0;
9444: if (strlen(modelsav) >1){
9445: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9446: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9447: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9448: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9449: * cst, age and age*age
9450: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9451: /* including age products which are counted in cptcovage.
9452: * but the covariates which are products must be treated
9453: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9454: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9455: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9456:
9457:
1.187 brouard 9458: /* Design
9459: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9460: * < ncovcol=8 >
9461: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9462: * k= 1 2 3 4 5 6 7 8
9463: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9464: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9465: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9466: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9467: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9468: * Tage[++cptcovage]=k
9469: * if products, new covar are created after ncovcol with k1
9470: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9471: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9472: * 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
9473: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9474: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9475: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9476: * < ncovcol=8 >
9477: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9478: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9479: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9480: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9481: * p Tprod[1]@2={ 6, 5}
9482: *p Tvard[1][1]@4= {7, 8, 5, 6}
9483: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9484: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9485: *How to reorganize?
9486: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9487: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9488: * {2, 1, 4, 8, 5, 6, 3, 7}
9489: * Struct []
9490: */
1.225 brouard 9491:
1.187 brouard 9492: /* This loop fills the array Tvar from the string 'model'.*/
9493: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9494: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9495: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9496: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9497: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9498: /* k=1 Tvar[1]=2 (from V2) */
9499: /* k=5 Tvar[5] */
9500: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9501: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9502: /* } */
1.198 brouard 9503: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9504: /*
9505: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9506: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9507: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9508: }
1.187 brouard 9509: cptcovage=0;
9510: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9511: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9512: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9513: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9514: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9515: /*scanf("%d",i);*/
9516: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9517: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9518: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9519: /* covar is not filled and then is empty */
9520: cptcovprod--;
9521: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9522: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9523: Typevar[k]=1; /* 1 for age product */
9524: cptcovage++; /* Sums the number of covariates which include age as a product */
9525: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9526: /*printf("stre=%s ", stre);*/
9527: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9528: cptcovprod--;
9529: cutl(stre,strb,strc,'V');
9530: Tvar[k]=atoi(stre);
9531: Typevar[k]=1; /* 1 for age product */
9532: cptcovage++;
9533: Tage[cptcovage]=k;
9534: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9535: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9536: cptcovn++;
9537: cptcovprodnoage++;k1++;
9538: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9539: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9540: because this model-covariate is a construction we invent a new column
9541: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9542: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9543: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9544: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9545: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9546: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9547: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9548: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9549: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9550: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9551: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9552: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9553: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9554: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9555: for (i=1; i<=lastobs;i++){
9556: /* Computes the new covariate which is a product of
9557: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9558: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9559: }
9560: } /* End age is not in the model */
9561: } /* End if model includes a product */
9562: else { /* no more sum */
9563: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9564: /* scanf("%d",i);*/
9565: cutl(strd,strc,strb,'V');
9566: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9567: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9568: Tvar[k]=atoi(strd);
9569: Typevar[k]=0; /* 0 for simple covariates */
9570: }
9571: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9572: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9573: scanf("%d",i);*/
1.187 brouard 9574: } /* end of loop + on total covariates */
9575: } /* end if strlen(modelsave == 0) age*age might exist */
9576: } /* end if strlen(model == 0) */
1.136 brouard 9577:
9578: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9579: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9580:
1.136 brouard 9581: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9582: printf("cptcovprod=%d ", cptcovprod);
9583: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9584: scanf("%d ",i);*/
9585:
9586:
1.230 brouard 9587: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9588: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9589: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9590: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9591: k = 1 2 3 4 5 6 7 8 9
9592: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9593: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9594: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9595: Dummy[k] 1 0 0 0 3 1 1 2 3
9596: Tmodelind[combination of covar]=k;
1.225 brouard 9597: */
9598: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9599: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9600: /* 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 9601: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9602: printf("Model=%s\n\
9603: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9604: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9605: 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);
9606: fprintf(ficlog,"Model=%s\n\
9607: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9608: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9609: 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 9610: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9611: 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 */
9612: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9613: Fixed[k]= 0;
9614: Dummy[k]= 0;
1.225 brouard 9615: ncoveff++;
1.232 brouard 9616: ncovf++;
1.234 brouard 9617: nsd++;
9618: modell[k].maintype= FTYPE;
9619: TvarsD[nsd]=Tvar[k];
9620: TvarsDind[nsd]=k;
9621: TvarF[ncovf]=Tvar[k];
9622: TvarFind[ncovf]=k;
9623: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9624: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9625: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9626: Fixed[k]= 0;
9627: Dummy[k]= 0;
9628: ncoveff++;
9629: ncovf++;
9630: modell[k].maintype= FTYPE;
9631: TvarF[ncovf]=Tvar[k];
9632: TvarFind[ncovf]=k;
1.230 brouard 9633: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9634: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9635: }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 9636: Fixed[k]= 0;
9637: Dummy[k]= 1;
1.230 brouard 9638: nqfveff++;
1.234 brouard 9639: modell[k].maintype= FTYPE;
9640: modell[k].subtype= FQ;
9641: nsq++;
9642: TvarsQ[nsq]=Tvar[k];
9643: TvarsQind[nsq]=k;
1.232 brouard 9644: ncovf++;
1.234 brouard 9645: TvarF[ncovf]=Tvar[k];
9646: TvarFind[ncovf]=k;
1.231 brouard 9647: 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 9648: 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 9649: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9650: Fixed[k]= 1;
9651: Dummy[k]= 0;
1.225 brouard 9652: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9653: modell[k].maintype= VTYPE;
9654: modell[k].subtype= VD;
9655: nsd++;
9656: TvarsD[nsd]=Tvar[k];
9657: TvarsDind[nsd]=k;
9658: ncovv++; /* Only simple time varying variables */
9659: TvarV[ncovv]=Tvar[k];
1.242 brouard 9660: 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 9661: 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 */
9662: 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 9663: 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);
9664: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9665: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9666: Fixed[k]= 1;
9667: Dummy[k]= 1;
9668: nqtveff++;
9669: modell[k].maintype= VTYPE;
9670: modell[k].subtype= VQ;
9671: ncovv++; /* Only simple time varying variables */
9672: nsq++;
9673: TvarsQ[nsq]=Tvar[k];
9674: TvarsQind[nsq]=k;
9675: TvarV[ncovv]=Tvar[k];
1.242 brouard 9676: 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 9677: 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 */
9678: 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 9679: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9680: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9681: 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 9682: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9683: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9684: ncova++;
9685: TvarA[ncova]=Tvar[k];
9686: TvarAind[ncova]=k;
1.231 brouard 9687: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9688: Fixed[k]= 2;
9689: Dummy[k]= 2;
9690: modell[k].maintype= ATYPE;
9691: modell[k].subtype= APFD;
9692: /* ncoveff++; */
1.227 brouard 9693: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9694: Fixed[k]= 2;
9695: Dummy[k]= 3;
9696: modell[k].maintype= ATYPE;
9697: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9698: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9699: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9700: Fixed[k]= 3;
9701: Dummy[k]= 2;
9702: modell[k].maintype= ATYPE;
9703: modell[k].subtype= APVD; /* Product age * varying dummy */
9704: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9705: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9706: Fixed[k]= 3;
9707: Dummy[k]= 3;
9708: modell[k].maintype= ATYPE;
9709: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9710: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9711: }
9712: }else if (Typevar[k] == 2) { /* product without age */
9713: k1=Tposprod[k];
9714: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9715: if(Tvard[k1][2] <=ncovcol){
9716: Fixed[k]= 1;
9717: Dummy[k]= 0;
9718: modell[k].maintype= FTYPE;
9719: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9720: ncovf++; /* Fixed variables without age */
9721: TvarF[ncovf]=Tvar[k];
9722: TvarFind[ncovf]=k;
9723: }else if(Tvard[k1][2] <=ncovcol+nqv){
9724: Fixed[k]= 0; /* or 2 ?*/
9725: Dummy[k]= 1;
9726: modell[k].maintype= FTYPE;
9727: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9728: ncovf++; /* Varying variables without age */
9729: TvarF[ncovf]=Tvar[k];
9730: TvarFind[ncovf]=k;
9731: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9732: Fixed[k]= 1;
9733: Dummy[k]= 0;
9734: modell[k].maintype= VTYPE;
9735: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9736: ncovv++; /* Varying variables without age */
9737: TvarV[ncovv]=Tvar[k];
9738: TvarVind[ncovv]=k;
9739: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9740: Fixed[k]= 1;
9741: Dummy[k]= 1;
9742: modell[k].maintype= VTYPE;
9743: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9744: ncovv++; /* Varying variables without age */
9745: TvarV[ncovv]=Tvar[k];
9746: TvarVind[ncovv]=k;
9747: }
1.227 brouard 9748: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9749: if(Tvard[k1][2] <=ncovcol){
9750: Fixed[k]= 0; /* or 2 ?*/
9751: Dummy[k]= 1;
9752: modell[k].maintype= FTYPE;
9753: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9754: ncovf++; /* Fixed variables without age */
9755: TvarF[ncovf]=Tvar[k];
9756: TvarFind[ncovf]=k;
9757: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9758: Fixed[k]= 1;
9759: Dummy[k]= 1;
9760: modell[k].maintype= VTYPE;
9761: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9762: ncovv++; /* Varying variables without age */
9763: TvarV[ncovv]=Tvar[k];
9764: TvarVind[ncovv]=k;
9765: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9766: Fixed[k]= 1;
9767: Dummy[k]= 1;
9768: modell[k].maintype= VTYPE;
9769: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9770: ncovv++; /* Varying variables without age */
9771: TvarV[ncovv]=Tvar[k];
9772: TvarVind[ncovv]=k;
9773: ncovv++; /* Varying variables without age */
9774: TvarV[ncovv]=Tvar[k];
9775: TvarVind[ncovv]=k;
9776: }
1.227 brouard 9777: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9778: if(Tvard[k1][2] <=ncovcol){
9779: Fixed[k]= 1;
9780: Dummy[k]= 1;
9781: modell[k].maintype= VTYPE;
9782: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9783: ncovv++; /* Varying variables without age */
9784: TvarV[ncovv]=Tvar[k];
9785: TvarVind[ncovv]=k;
9786: }else if(Tvard[k1][2] <=ncovcol+nqv){
9787: Fixed[k]= 1;
9788: Dummy[k]= 1;
9789: modell[k].maintype= VTYPE;
9790: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9791: ncovv++; /* Varying variables without age */
9792: TvarV[ncovv]=Tvar[k];
9793: TvarVind[ncovv]=k;
9794: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9795: Fixed[k]= 1;
9796: Dummy[k]= 0;
9797: modell[k].maintype= VTYPE;
9798: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9799: ncovv++; /* Varying variables without age */
9800: TvarV[ncovv]=Tvar[k];
9801: TvarVind[ncovv]=k;
9802: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9803: Fixed[k]= 1;
9804: Dummy[k]= 1;
9805: modell[k].maintype= VTYPE;
9806: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9807: ncovv++; /* Varying variables without age */
9808: TvarV[ncovv]=Tvar[k];
9809: TvarVind[ncovv]=k;
9810: }
1.227 brouard 9811: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9812: if(Tvard[k1][2] <=ncovcol){
9813: Fixed[k]= 1;
9814: Dummy[k]= 1;
9815: modell[k].maintype= VTYPE;
9816: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9817: ncovv++; /* Varying variables without age */
9818: TvarV[ncovv]=Tvar[k];
9819: TvarVind[ncovv]=k;
9820: }else if(Tvard[k1][2] <=ncovcol+nqv){
9821: Fixed[k]= 1;
9822: Dummy[k]= 1;
9823: modell[k].maintype= VTYPE;
9824: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9825: ncovv++; /* Varying variables without age */
9826: TvarV[ncovv]=Tvar[k];
9827: TvarVind[ncovv]=k;
9828: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9829: Fixed[k]= 1;
9830: Dummy[k]= 1;
9831: modell[k].maintype= VTYPE;
9832: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9833: ncovv++; /* Varying variables without age */
9834: TvarV[ncovv]=Tvar[k];
9835: TvarVind[ncovv]=k;
9836: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9837: Fixed[k]= 1;
9838: Dummy[k]= 1;
9839: modell[k].maintype= VTYPE;
9840: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9841: ncovv++; /* Varying variables without age */
9842: TvarV[ncovv]=Tvar[k];
9843: TvarVind[ncovv]=k;
9844: }
1.227 brouard 9845: }else{
1.240 brouard 9846: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9847: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9848: } /*end k1*/
1.225 brouard 9849: }else{
1.226 brouard 9850: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9851: 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 9852: }
1.227 brouard 9853: 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 9854: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9855: 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]);
9856: }
9857: /* Searching for doublons in the model */
9858: for(k1=1; k1<= cptcovt;k1++){
9859: for(k2=1; k2 <k1;k2++){
9860: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9861: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9862: if(Tvar[k1]==Tvar[k2]){
9863: 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]]);
9864: 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);
9865: return(1);
9866: }
9867: }else if (Typevar[k1] ==2){
9868: k3=Tposprod[k1];
9869: k4=Tposprod[k2];
9870: 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])) ){
9871: 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]]);
9872: 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);
9873: return(1);
9874: }
9875: }
1.227 brouard 9876: }
9877: }
1.225 brouard 9878: }
9879: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9880: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9881: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9882: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9883: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9884: /*endread:*/
1.225 brouard 9885: printf("Exiting decodemodel: ");
9886: return (1);
1.136 brouard 9887: }
9888:
1.169 brouard 9889: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9890: {/* Check ages at death */
1.136 brouard 9891: int i, m;
1.218 brouard 9892: int firstone=0;
9893:
1.136 brouard 9894: for (i=1; i<=imx; i++) {
9895: for(m=2; (m<= maxwav); m++) {
9896: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9897: anint[m][i]=9999;
1.216 brouard 9898: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9899: s[m][i]=-1;
1.136 brouard 9900: }
9901: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9902: *nberr = *nberr + 1;
1.218 brouard 9903: if(firstone == 0){
9904: firstone=1;
1.260 brouard 9905: printf("Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 9906: }
1.262 brouard 9907: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.260 brouard 9908: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9909: }
9910: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9911: (*nberr)++;
1.259 brouard 9912: printf("Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.262 brouard 9913: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.259 brouard 9914: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9915: }
9916: }
9917: }
9918:
9919: for (i=1; i<=imx; i++) {
9920: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9921: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9922: 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 9923: if (s[m][i] >= nlstate+1) {
1.169 brouard 9924: if(agedc[i]>0){
9925: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9926: agev[m][i]=agedc[i];
1.214 brouard 9927: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9928: }else {
1.136 brouard 9929: if ((int)andc[i]!=9999){
9930: nbwarn++;
9931: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9932: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9933: agev[m][i]=-1;
9934: }
9935: }
1.169 brouard 9936: } /* agedc > 0 */
1.214 brouard 9937: } /* end if */
1.136 brouard 9938: else if(s[m][i] !=9){ /* Standard case, age in fractional
9939: years but with the precision of a month */
9940: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9941: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9942: agev[m][i]=1;
9943: else if(agev[m][i] < *agemin){
9944: *agemin=agev[m][i];
9945: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9946: }
9947: else if(agev[m][i] >*agemax){
9948: *agemax=agev[m][i];
1.156 brouard 9949: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9950: }
9951: /*agev[m][i]=anint[m][i]-annais[i];*/
9952: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9953: } /* en if 9*/
1.136 brouard 9954: else { /* =9 */
1.214 brouard 9955: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9956: agev[m][i]=1;
9957: s[m][i]=-1;
9958: }
9959: }
1.214 brouard 9960: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9961: agev[m][i]=1;
1.214 brouard 9962: else{
9963: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9964: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9965: agev[m][i]=0;
9966: }
9967: } /* End for lastpass */
9968: }
1.136 brouard 9969:
9970: for (i=1; i<=imx; i++) {
9971: for(m=firstpass; (m<=lastpass); m++){
9972: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9973: (*nberr)++;
1.136 brouard 9974: 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);
9975: 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);
9976: return 1;
9977: }
9978: }
9979: }
9980:
9981: /*for (i=1; i<=imx; i++){
9982: for (m=firstpass; (m<lastpass); m++){
9983: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9984: }
9985:
9986: }*/
9987:
9988:
1.139 brouard 9989: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9990: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9991:
9992: return (0);
1.164 brouard 9993: /* endread:*/
1.136 brouard 9994: printf("Exiting calandcheckages: ");
9995: return (1);
9996: }
9997:
1.172 brouard 9998: #if defined(_MSC_VER)
9999: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10000: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10001: //#include "stdafx.h"
10002: //#include <stdio.h>
10003: //#include <tchar.h>
10004: //#include <windows.h>
10005: //#include <iostream>
10006: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10007:
10008: LPFN_ISWOW64PROCESS fnIsWow64Process;
10009:
10010: BOOL IsWow64()
10011: {
10012: BOOL bIsWow64 = FALSE;
10013:
10014: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10015: // (HANDLE, PBOOL);
10016:
10017: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10018:
10019: HMODULE module = GetModuleHandle(_T("kernel32"));
10020: const char funcName[] = "IsWow64Process";
10021: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10022: GetProcAddress(module, funcName);
10023:
10024: if (NULL != fnIsWow64Process)
10025: {
10026: if (!fnIsWow64Process(GetCurrentProcess(),
10027: &bIsWow64))
10028: //throw std::exception("Unknown error");
10029: printf("Unknown error\n");
10030: }
10031: return bIsWow64 != FALSE;
10032: }
10033: #endif
1.177 brouard 10034:
1.191 brouard 10035: void syscompilerinfo(int logged)
1.167 brouard 10036: {
10037: /* #include "syscompilerinfo.h"*/
1.185 brouard 10038: /* command line Intel compiler 32bit windows, XP compatible:*/
10039: /* /GS /W3 /Gy
10040: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10041: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10042: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10043: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10044: */
10045: /* 64 bits */
1.185 brouard 10046: /*
10047: /GS /W3 /Gy
10048: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10049: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10050: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10051: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10052: /* Optimization are useless and O3 is slower than O2 */
10053: /*
10054: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10055: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10056: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10057: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10058: */
1.186 brouard 10059: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10060: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10061: /PDB:"visual studio
10062: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10063: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10064: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10065: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10066: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10067: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10068: uiAccess='false'"
10069: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10070: /NOLOGO /TLBID:1
10071: */
1.177 brouard 10072: #if defined __INTEL_COMPILER
1.178 brouard 10073: #if defined(__GNUC__)
10074: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10075: #endif
1.177 brouard 10076: #elif defined(__GNUC__)
1.179 brouard 10077: #ifndef __APPLE__
1.174 brouard 10078: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10079: #endif
1.177 brouard 10080: struct utsname sysInfo;
1.178 brouard 10081: int cross = CROSS;
10082: if (cross){
10083: printf("Cross-");
1.191 brouard 10084: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10085: }
1.174 brouard 10086: #endif
10087:
1.171 brouard 10088: #include <stdint.h>
1.178 brouard 10089:
1.191 brouard 10090: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10091: #if defined(__clang__)
1.191 brouard 10092: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10093: #endif
10094: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10095: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10096: #endif
10097: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10098: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10099: #endif
10100: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10101: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10102: #endif
10103: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10104: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10105: #endif
10106: #if defined(_MSC_VER)
1.191 brouard 10107: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10108: #endif
10109: #if defined(__PGI)
1.191 brouard 10110: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10111: #endif
10112: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10113: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10114: #endif
1.191 brouard 10115: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10116:
1.167 brouard 10117: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10118: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10119: // Windows (x64 and x86)
1.191 brouard 10120: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10121: #elif __unix__ // all unices, not all compilers
10122: // Unix
1.191 brouard 10123: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10124: #elif __linux__
10125: // linux
1.191 brouard 10126: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10127: #elif __APPLE__
1.174 brouard 10128: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10129: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10130: #endif
10131:
10132: /* __MINGW32__ */
10133: /* __CYGWIN__ */
10134: /* __MINGW64__ */
10135: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10136: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10137: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10138: /* _WIN64 // Defined for applications for Win64. */
10139: /* _M_X64 // Defined for compilations that target x64 processors. */
10140: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10141:
1.167 brouard 10142: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10143: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10144: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10145: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10146: #else
1.191 brouard 10147: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10148: #endif
10149:
1.169 brouard 10150: #if defined(__GNUC__)
10151: # if defined(__GNUC_PATCHLEVEL__)
10152: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10153: + __GNUC_MINOR__ * 100 \
10154: + __GNUC_PATCHLEVEL__)
10155: # else
10156: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10157: + __GNUC_MINOR__ * 100)
10158: # endif
1.174 brouard 10159: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10160: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10161:
10162: if (uname(&sysInfo) != -1) {
10163: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10164: 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 10165: }
10166: else
10167: perror("uname() error");
1.179 brouard 10168: //#ifndef __INTEL_COMPILER
10169: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10170: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10171: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10172: #endif
1.169 brouard 10173: #endif
1.172 brouard 10174:
10175: // void main()
10176: // {
1.169 brouard 10177: #if defined(_MSC_VER)
1.174 brouard 10178: if (IsWow64()){
1.191 brouard 10179: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10180: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10181: }
10182: else{
1.191 brouard 10183: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10184: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10185: }
1.172 brouard 10186: // printf("\nPress Enter to continue...");
10187: // getchar();
10188: // }
10189:
1.169 brouard 10190: #endif
10191:
1.167 brouard 10192:
1.219 brouard 10193: }
1.136 brouard 10194:
1.219 brouard 10195: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10196: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10197: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10198: /* double ftolpl = 1.e-10; */
1.180 brouard 10199: double age, agebase, agelim;
1.203 brouard 10200: double tot;
1.180 brouard 10201:
1.202 brouard 10202: strcpy(filerespl,"PL_");
10203: strcat(filerespl,fileresu);
10204: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10205: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10206: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10207: }
1.227 brouard 10208: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10209: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10210: pstamp(ficrespl);
1.203 brouard 10211: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10212: fprintf(ficrespl,"#Age ");
10213: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10214: fprintf(ficrespl,"\n");
1.180 brouard 10215:
1.219 brouard 10216: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10217:
1.219 brouard 10218: agebase=ageminpar;
10219: agelim=agemaxpar;
1.180 brouard 10220:
1.227 brouard 10221: /* i1=pow(2,ncoveff); */
1.234 brouard 10222: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10223: if (cptcovn < 1){i1=1;}
1.180 brouard 10224:
1.238 brouard 10225: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10226: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10227: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10228: continue;
1.235 brouard 10229:
1.238 brouard 10230: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10231: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10232: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10233: /* k=k+1; */
10234: /* to clean */
10235: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10236: fprintf(ficrespl,"#******");
10237: printf("#******");
10238: fprintf(ficlog,"#******");
10239: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10240: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10241: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10242: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10243: }
10244: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10245: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10246: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10247: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10248: }
10249: fprintf(ficrespl,"******\n");
10250: printf("******\n");
10251: fprintf(ficlog,"******\n");
10252: if(invalidvarcomb[k]){
10253: printf("\nCombination (%d) ignored because no case \n",k);
10254: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10255: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10256: continue;
10257: }
1.219 brouard 10258:
1.238 brouard 10259: fprintf(ficrespl,"#Age ");
10260: for(j=1;j<=cptcoveff;j++) {
10261: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10262: }
10263: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10264: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10265:
1.238 brouard 10266: for (age=agebase; age<=agelim; age++){
10267: /* for (age=agebase; age<=agebase; age++){ */
10268: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10269: fprintf(ficrespl,"%.0f ",age );
10270: for(j=1;j<=cptcoveff;j++)
10271: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10272: tot=0.;
10273: for(i=1; i<=nlstate;i++){
10274: tot += prlim[i][i];
10275: fprintf(ficrespl," %.5f", prlim[i][i]);
10276: }
10277: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10278: } /* Age */
10279: /* was end of cptcod */
10280: } /* cptcov */
10281: } /* nres */
1.219 brouard 10282: return 0;
1.180 brouard 10283: }
10284:
1.218 brouard 10285: 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){
10286: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10287:
10288: /* Computes the back prevalence limit for any combination of covariate values
10289: * at any age between ageminpar and agemaxpar
10290: */
1.235 brouard 10291: int i, j, k, i1, nres=0 ;
1.217 brouard 10292: /* double ftolpl = 1.e-10; */
10293: double age, agebase, agelim;
10294: double tot;
1.218 brouard 10295: /* double ***mobaverage; */
10296: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10297:
10298: strcpy(fileresplb,"PLB_");
10299: strcat(fileresplb,fileresu);
10300: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10301: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10302: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10303: }
10304: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10305: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10306: pstamp(ficresplb);
10307: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10308: fprintf(ficresplb,"#Age ");
10309: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10310: fprintf(ficresplb,"\n");
10311:
1.218 brouard 10312:
10313: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10314:
10315: agebase=ageminpar;
10316: agelim=agemaxpar;
10317:
10318:
1.227 brouard 10319: i1=pow(2,cptcoveff);
1.218 brouard 10320: if (cptcovn < 1){i1=1;}
1.227 brouard 10321:
1.238 brouard 10322: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10323: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10324: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10325: continue;
10326: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10327: fprintf(ficresplb,"#******");
10328: printf("#******");
10329: fprintf(ficlog,"#******");
10330: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10331: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10332: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10333: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10334: }
10335: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10336: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10337: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10338: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10339: }
10340: fprintf(ficresplb,"******\n");
10341: printf("******\n");
10342: fprintf(ficlog,"******\n");
10343: if(invalidvarcomb[k]){
10344: printf("\nCombination (%d) ignored because no cases \n",k);
10345: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10346: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10347: continue;
10348: }
1.218 brouard 10349:
1.238 brouard 10350: fprintf(ficresplb,"#Age ");
10351: for(j=1;j<=cptcoveff;j++) {
10352: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10353: }
10354: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10355: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10356:
10357:
1.238 brouard 10358: for (age=agebase; age<=agelim; age++){
10359: /* for (age=agebase; age<=agebase; age++){ */
10360: if(mobilavproj > 0){
10361: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10362: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10363: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10364: }else if (mobilavproj == 0){
10365: 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);
10366: 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);
10367: exit(1);
10368: }else{
10369: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10370: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10371: /* printf("TOTOT\n"); */
10372: /* exit(1); */
1.238 brouard 10373: }
10374: fprintf(ficresplb,"%.0f ",age );
10375: for(j=1;j<=cptcoveff;j++)
10376: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10377: tot=0.;
10378: for(i=1; i<=nlstate;i++){
10379: tot += bprlim[i][i];
10380: fprintf(ficresplb," %.5f", bprlim[i][i]);
10381: }
10382: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10383: } /* Age */
10384: /* was end of cptcod */
1.255 brouard 10385: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10386: } /* end of any combination */
10387: } /* end of nres */
1.218 brouard 10388: /* hBijx(p, bage, fage); */
10389: /* fclose(ficrespijb); */
10390:
10391: return 0;
1.217 brouard 10392: }
1.218 brouard 10393:
1.180 brouard 10394: int hPijx(double *p, int bage, int fage){
10395: /*------------- h Pij x at various ages ------------*/
10396:
10397: int stepsize;
10398: int agelim;
10399: int hstepm;
10400: int nhstepm;
1.235 brouard 10401: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10402:
10403: double agedeb;
10404: double ***p3mat;
10405:
1.201 brouard 10406: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10407: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10408: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10409: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10410: }
10411: printf("Computing pij: result on file '%s' \n", filerespij);
10412: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10413:
10414: stepsize=(int) (stepm+YEARM-1)/YEARM;
10415: /*if (stepm<=24) stepsize=2;*/
10416:
10417: agelim=AGESUP;
10418: hstepm=stepsize*YEARM; /* Every year of age */
10419: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10420:
1.180 brouard 10421: /* hstepm=1; aff par mois*/
10422: pstamp(ficrespij);
10423: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10424: i1= pow(2,cptcoveff);
1.218 brouard 10425: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10426: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10427: /* k=k+1; */
1.235 brouard 10428: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10429: for(k=1; k<=i1;k++){
1.253 brouard 10430: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10431: continue;
1.183 brouard 10432: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10433: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10434: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10435: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10436: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10437: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10438: }
1.183 brouard 10439: fprintf(ficrespij,"******\n");
10440:
10441: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10442: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10443: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10444:
10445: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10446:
1.183 brouard 10447: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10448: oldm=oldms;savm=savms;
1.235 brouard 10449: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10450: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10451: for(i=1; i<=nlstate;i++)
10452: for(j=1; j<=nlstate+ndeath;j++)
10453: fprintf(ficrespij," %1d-%1d",i,j);
10454: fprintf(ficrespij,"\n");
10455: for (h=0; h<=nhstepm; h++){
10456: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10457: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10458: for(i=1; i<=nlstate;i++)
10459: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10460: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10461: fprintf(ficrespij,"\n");
10462: }
1.183 brouard 10463: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10464: fprintf(ficrespij,"\n");
10465: }
1.180 brouard 10466: /*}*/
10467: }
1.218 brouard 10468: return 0;
1.180 brouard 10469: }
1.218 brouard 10470:
10471: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10472: /*------------- h Bij x at various ages ------------*/
10473:
10474: int stepsize;
1.218 brouard 10475: /* int agelim; */
10476: int ageminl;
1.217 brouard 10477: int hstepm;
10478: int nhstepm;
1.238 brouard 10479: int h, i, i1, j, k, nres;
1.218 brouard 10480:
1.217 brouard 10481: double agedeb;
10482: double ***p3mat;
1.218 brouard 10483:
10484: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10485: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10486: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10487: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10488: }
10489: printf("Computing pij back: result on file '%s' \n", filerespijb);
10490: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10491:
10492: stepsize=(int) (stepm+YEARM-1)/YEARM;
10493: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10494:
1.218 brouard 10495: /* agelim=AGESUP; */
10496: ageminl=30;
10497: hstepm=stepsize*YEARM; /* Every year of age */
10498: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10499:
10500: /* hstepm=1; aff par mois*/
10501: pstamp(ficrespijb);
1.255 brouard 10502: 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 10503: i1= pow(2,cptcoveff);
1.218 brouard 10504: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10505: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10506: /* k=k+1; */
1.238 brouard 10507: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10508: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10509: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10510: continue;
10511: fprintf(ficrespijb,"\n#****** ");
10512: for(j=1;j<=cptcoveff;j++)
10513: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10514: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10515: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10516: }
10517: fprintf(ficrespijb,"******\n");
1.264 brouard 10518: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10519: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10520: continue;
10521: }
10522:
10523: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10524: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10525: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10526: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10527: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10528:
10529: /* nhstepm=nhstepm*YEARM; aff par mois*/
10530:
1.266 brouard 10531: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10532: /* and memory limitations if stepm is small */
10533:
1.238 brouard 10534: /* oldm=oldms;savm=savms; */
10535: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10536: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10537: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10538: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10539: for(i=1; i<=nlstate;i++)
10540: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10541: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10542: fprintf(ficrespijb,"\n");
1.238 brouard 10543: for (h=0; h<=nhstepm; h++){
10544: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10545: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10546: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10547: for(i=1; i<=nlstate;i++)
10548: for(j=1; j<=nlstate+ndeath;j++)
10549: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10550: fprintf(ficrespijb,"\n");
10551: }
10552: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10553: fprintf(ficrespijb,"\n");
10554: } /* end age deb */
10555: } /* end combination */
10556: } /* end nres */
1.218 brouard 10557: return 0;
10558: } /* hBijx */
1.217 brouard 10559:
1.180 brouard 10560:
1.136 brouard 10561: /***********************************************/
10562: /**************** Main Program *****************/
10563: /***********************************************/
10564:
10565: int main(int argc, char *argv[])
10566: {
10567: #ifdef GSL
10568: const gsl_multimin_fminimizer_type *T;
10569: size_t iteri = 0, it;
10570: int rval = GSL_CONTINUE;
10571: int status = GSL_SUCCESS;
10572: double ssval;
10573: #endif
10574: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10575: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10576: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10577: int jj, ll, li, lj, lk;
1.136 brouard 10578: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10579: int num_filled;
1.136 brouard 10580: int itimes;
10581: int NDIM=2;
10582: int vpopbased=0;
1.235 brouard 10583: int nres=0;
1.258 brouard 10584: int endishere=0;
1.136 brouard 10585:
1.164 brouard 10586: char ca[32], cb[32];
1.136 brouard 10587: /* FILE *fichtm; *//* Html File */
10588: /* FILE *ficgp;*/ /*Gnuplot File */
10589: struct stat info;
1.191 brouard 10590: double agedeb=0.;
1.194 brouard 10591:
10592: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10593: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10594:
1.165 brouard 10595: double fret;
1.191 brouard 10596: double dum=0.; /* Dummy variable */
1.136 brouard 10597: double ***p3mat;
1.218 brouard 10598: /* double ***mobaverage; */
1.164 brouard 10599:
10600: char line[MAXLINE];
1.197 brouard 10601: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10602:
1.234 brouard 10603: char modeltemp[MAXLINE];
1.230 brouard 10604: char resultline[MAXLINE];
10605:
1.136 brouard 10606: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10607: char *tok, *val; /* pathtot */
1.136 brouard 10608: int firstobs=1, lastobs=10;
1.195 brouard 10609: int c, h , cpt, c2;
1.191 brouard 10610: int jl=0;
10611: int i1, j1, jk, stepsize=0;
1.194 brouard 10612: int count=0;
10613:
1.164 brouard 10614: int *tab;
1.136 brouard 10615: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10616: int backcast=0;
1.136 brouard 10617: int mobilav=0,popforecast=0;
1.191 brouard 10618: int hstepm=0, nhstepm=0;
1.136 brouard 10619: int agemortsup;
10620: float sumlpop=0.;
10621: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10622: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10623:
1.191 brouard 10624: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10625: double ftolpl=FTOL;
10626: double **prlim;
1.217 brouard 10627: double **bprlim;
1.136 brouard 10628: double ***param; /* Matrix of parameters */
1.251 brouard 10629: double ***paramstart; /* Matrix of starting parameter values */
10630: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10631: double **matcov; /* Matrix of covariance */
1.203 brouard 10632: double **hess; /* Hessian matrix */
1.136 brouard 10633: double ***delti3; /* Scale */
10634: double *delti; /* Scale */
10635: double ***eij, ***vareij;
10636: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10637:
1.136 brouard 10638: double *epj, vepp;
1.164 brouard 10639:
1.273 ! brouard 10640: double dateprev1, dateprev2;
! 10641: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
! 10642: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10643:
1.136 brouard 10644: double **ximort;
1.145 brouard 10645: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10646: int *dcwave;
10647:
1.164 brouard 10648: char z[1]="c";
1.136 brouard 10649:
10650: /*char *strt;*/
10651: char strtend[80];
1.126 brouard 10652:
1.164 brouard 10653:
1.126 brouard 10654: /* setlocale (LC_ALL, ""); */
10655: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10656: /* textdomain (PACKAGE); */
10657: /* setlocale (LC_CTYPE, ""); */
10658: /* setlocale (LC_MESSAGES, ""); */
10659:
10660: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10661: rstart_time = time(NULL);
10662: /* (void) gettimeofday(&start_time,&tzp);*/
10663: start_time = *localtime(&rstart_time);
1.126 brouard 10664: curr_time=start_time;
1.157 brouard 10665: /*tml = *localtime(&start_time.tm_sec);*/
10666: /* strcpy(strstart,asctime(&tml)); */
10667: strcpy(strstart,asctime(&start_time));
1.126 brouard 10668:
10669: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10670: /* tp.tm_sec = tp.tm_sec +86400; */
10671: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10672: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10673: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10674: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10675: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10676: /* strt=asctime(&tmg); */
10677: /* printf("Time(after) =%s",strstart); */
10678: /* (void) time (&time_value);
10679: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10680: * tm = *localtime(&time_value);
10681: * strstart=asctime(&tm);
10682: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10683: */
10684:
10685: nberr=0; /* Number of errors and warnings */
10686: nbwarn=0;
1.184 brouard 10687: #ifdef WIN32
10688: _getcwd(pathcd, size);
10689: #else
1.126 brouard 10690: getcwd(pathcd, size);
1.184 brouard 10691: #endif
1.191 brouard 10692: syscompilerinfo(0);
1.196 brouard 10693: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10694: if(argc <=1){
10695: printf("\nEnter the parameter file name: ");
1.205 brouard 10696: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10697: printf("ERROR Empty parameter file name\n");
10698: goto end;
10699: }
1.126 brouard 10700: i=strlen(pathr);
10701: if(pathr[i-1]=='\n')
10702: pathr[i-1]='\0';
1.156 brouard 10703: i=strlen(pathr);
1.205 brouard 10704: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10705: pathr[i-1]='\0';
1.205 brouard 10706: }
10707: i=strlen(pathr);
10708: if( i==0 ){
10709: printf("ERROR Empty parameter file name\n");
10710: goto end;
10711: }
10712: for (tok = pathr; tok != NULL; ){
1.126 brouard 10713: printf("Pathr |%s|\n",pathr);
10714: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10715: printf("val= |%s| pathr=%s\n",val,pathr);
10716: strcpy (pathtot, val);
10717: if(pathr[0] == '\0') break; /* Dirty */
10718: }
10719: }
10720: else{
10721: strcpy(pathtot,argv[1]);
10722: }
10723: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10724: /*cygwin_split_path(pathtot,path,optionfile);
10725: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10726: /* cutv(path,optionfile,pathtot,'\\');*/
10727:
10728: /* Split argv[0], imach program to get pathimach */
10729: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10730: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10731: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10732: /* strcpy(pathimach,argv[0]); */
10733: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10734: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10735: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10736: #ifdef WIN32
10737: _chdir(path); /* Can be a relative path */
10738: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10739: #else
1.126 brouard 10740: chdir(path); /* Can be a relative path */
1.184 brouard 10741: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10742: #endif
10743: printf("Current directory %s!\n",pathcd);
1.126 brouard 10744: strcpy(command,"mkdir ");
10745: strcat(command,optionfilefiname);
10746: if((outcmd=system(command)) != 0){
1.169 brouard 10747: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10748: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10749: /* fclose(ficlog); */
10750: /* exit(1); */
10751: }
10752: /* if((imk=mkdir(optionfilefiname))<0){ */
10753: /* perror("mkdir"); */
10754: /* } */
10755:
10756: /*-------- arguments in the command line --------*/
10757:
1.186 brouard 10758: /* Main Log file */
1.126 brouard 10759: strcat(filelog, optionfilefiname);
10760: strcat(filelog,".log"); /* */
10761: if((ficlog=fopen(filelog,"w"))==NULL) {
10762: printf("Problem with logfile %s\n",filelog);
10763: goto end;
10764: }
10765: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10766: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10767: fprintf(ficlog,"\nEnter the parameter file name: \n");
10768: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10769: path=%s \n\
10770: optionfile=%s\n\
10771: optionfilext=%s\n\
1.156 brouard 10772: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10773:
1.197 brouard 10774: syscompilerinfo(1);
1.167 brouard 10775:
1.126 brouard 10776: printf("Local time (at start):%s",strstart);
10777: fprintf(ficlog,"Local time (at start): %s",strstart);
10778: fflush(ficlog);
10779: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10780: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10781:
10782: /* */
10783: strcpy(fileres,"r");
10784: strcat(fileres, optionfilefiname);
1.201 brouard 10785: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10786: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10787: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10788:
1.186 brouard 10789: /* Main ---------arguments file --------*/
1.126 brouard 10790:
10791: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10792: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10793: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10794: fflush(ficlog);
1.149 brouard 10795: /* goto end; */
10796: exit(70);
1.126 brouard 10797: }
10798:
10799:
10800:
10801: strcpy(filereso,"o");
1.201 brouard 10802: strcat(filereso,fileresu);
1.126 brouard 10803: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10804: printf("Problem with Output resultfile: %s\n", filereso);
10805: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10806: fflush(ficlog);
10807: goto end;
10808: }
10809:
10810: /* Reads comments: lines beginning with '#' */
10811: numlinepar=0;
1.197 brouard 10812:
10813: /* First parameter line */
10814: while(fgets(line, MAXLINE, ficpar)) {
10815: /* If line starts with a # it is a comment */
10816: if (line[0] == '#') {
10817: numlinepar++;
10818: fputs(line,stdout);
10819: fputs(line,ficparo);
10820: fputs(line,ficlog);
10821: continue;
10822: }else
10823: break;
10824: }
10825: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10826: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10827: if (num_filled != 5) {
10828: printf("Should be 5 parameters\n");
10829: }
1.126 brouard 10830: numlinepar++;
1.197 brouard 10831: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10832: }
10833: /* Second parameter line */
10834: while(fgets(line, MAXLINE, ficpar)) {
10835: /* If line starts with a # it is a comment */
10836: if (line[0] == '#') {
10837: numlinepar++;
10838: fputs(line,stdout);
10839: fputs(line,ficparo);
10840: fputs(line,ficlog);
10841: continue;
10842: }else
10843: break;
10844: }
1.223 brouard 10845: 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", \
10846: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10847: if (num_filled != 11) {
10848: 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 10849: printf("but line=%s\n",line);
1.197 brouard 10850: }
1.223 brouard 10851: 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 10852: }
1.203 brouard 10853: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10854: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10855: /* Third parameter line */
10856: while(fgets(line, MAXLINE, ficpar)) {
10857: /* If line starts with a # it is a comment */
10858: if (line[0] == '#') {
10859: numlinepar++;
10860: fputs(line,stdout);
10861: fputs(line,ficparo);
10862: fputs(line,ficlog);
10863: continue;
10864: }else
10865: break;
10866: }
1.201 brouard 10867: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10868: if (num_filled == 0){
10869: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10870: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10871: model[0]='\0';
10872: goto end;
10873: } else if (num_filled != 1){
1.197 brouard 10874: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10875: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10876: model[0]='\0';
10877: goto end;
10878: }
10879: else{
10880: if (model[0]=='+'){
10881: for(i=1; i<=strlen(model);i++)
10882: modeltemp[i-1]=model[i];
1.201 brouard 10883: strcpy(model,modeltemp);
1.197 brouard 10884: }
10885: }
1.199 brouard 10886: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10887: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10888: }
10889: /* 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); */
10890: /* numlinepar=numlinepar+3; /\* In general *\/ */
10891: /* 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 10892: 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);
10893: 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 10894: fflush(ficlog);
1.190 brouard 10895: /* if(model[0]=='#'|| model[0]== '\0'){ */
10896: if(model[0]=='#'){
1.187 brouard 10897: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10898: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10899: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10900: if(mle != -1){
10901: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10902: exit(1);
10903: }
10904: }
1.126 brouard 10905: while((c=getc(ficpar))=='#' && c!= EOF){
10906: ungetc(c,ficpar);
10907: fgets(line, MAXLINE, ficpar);
10908: numlinepar++;
1.195 brouard 10909: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10910: z[0]=line[1];
10911: }
10912: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10913: fputs(line, stdout);
10914: //puts(line);
1.126 brouard 10915: fputs(line,ficparo);
10916: fputs(line,ficlog);
10917: }
10918: ungetc(c,ficpar);
10919:
10920:
1.145 brouard 10921: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10922: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10923: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10924: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10925: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10926: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10927: v1+v2*age+v2*v3 makes cptcovn = 3
10928: */
10929: if (strlen(model)>1)
1.187 brouard 10930: 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 10931: else
1.187 brouard 10932: ncovmodel=2; /* Constant and age */
1.133 brouard 10933: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10934: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10935: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10936: 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);
10937: 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);
10938: fflush(stdout);
10939: fclose (ficlog);
10940: goto end;
10941: }
1.126 brouard 10942: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10943: delti=delti3[1][1];
10944: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10945: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10946: /* We could also provide initial parameters values giving by simple logistic regression
10947: * only one way, that is without matrix product. We will have nlstate maximizations */
10948: /* for(i=1;i<nlstate;i++){ */
10949: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10950: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10951: /* } */
1.126 brouard 10952: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10953: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10954: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10955: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10956: fclose (ficparo);
10957: fclose (ficlog);
10958: goto end;
10959: exit(0);
1.220 brouard 10960: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10961: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10962: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10963: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10964: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10965: matcov=matrix(1,npar,1,npar);
1.203 brouard 10966: hess=matrix(1,npar,1,npar);
1.220 brouard 10967: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10968: /* Read guessed parameters */
1.126 brouard 10969: /* Reads comments: lines beginning with '#' */
10970: while((c=getc(ficpar))=='#' && c!= EOF){
10971: ungetc(c,ficpar);
10972: fgets(line, MAXLINE, ficpar);
10973: numlinepar++;
1.141 brouard 10974: fputs(line,stdout);
1.126 brouard 10975: fputs(line,ficparo);
10976: fputs(line,ficlog);
10977: }
10978: ungetc(c,ficpar);
10979:
10980: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10981: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10982: for(i=1; i <=nlstate; i++){
1.234 brouard 10983: j=0;
1.126 brouard 10984: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10985: if(jj==i) continue;
10986: j++;
10987: fscanf(ficpar,"%1d%1d",&i1,&j1);
10988: if ((i1 != i) || (j1 != jj)){
10989: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10990: It might be a problem of design; if ncovcol and the model are correct\n \
10991: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10992: exit(1);
10993: }
10994: fprintf(ficparo,"%1d%1d",i1,j1);
10995: if(mle==1)
10996: printf("%1d%1d",i,jj);
10997: fprintf(ficlog,"%1d%1d",i,jj);
10998: for(k=1; k<=ncovmodel;k++){
10999: fscanf(ficpar," %lf",¶m[i][j][k]);
11000: if(mle==1){
11001: printf(" %lf",param[i][j][k]);
11002: fprintf(ficlog," %lf",param[i][j][k]);
11003: }
11004: else
11005: fprintf(ficlog," %lf",param[i][j][k]);
11006: fprintf(ficparo," %lf",param[i][j][k]);
11007: }
11008: fscanf(ficpar,"\n");
11009: numlinepar++;
11010: if(mle==1)
11011: printf("\n");
11012: fprintf(ficlog,"\n");
11013: fprintf(ficparo,"\n");
1.126 brouard 11014: }
11015: }
11016: fflush(ficlog);
1.234 brouard 11017:
1.251 brouard 11018: /* Reads parameters values */
1.126 brouard 11019: p=param[1][1];
1.251 brouard 11020: pstart=paramstart[1][1];
1.126 brouard 11021:
11022: /* Reads comments: lines beginning with '#' */
11023: while((c=getc(ficpar))=='#' && c!= EOF){
11024: ungetc(c,ficpar);
11025: fgets(line, MAXLINE, ficpar);
11026: numlinepar++;
1.141 brouard 11027: fputs(line,stdout);
1.126 brouard 11028: fputs(line,ficparo);
11029: fputs(line,ficlog);
11030: }
11031: ungetc(c,ficpar);
11032:
11033: for(i=1; i <=nlstate; i++){
11034: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11035: fscanf(ficpar,"%1d%1d",&i1,&j1);
11036: if ( (i1-i) * (j1-j) != 0){
11037: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11038: exit(1);
11039: }
11040: printf("%1d%1d",i,j);
11041: fprintf(ficparo,"%1d%1d",i1,j1);
11042: fprintf(ficlog,"%1d%1d",i1,j1);
11043: for(k=1; k<=ncovmodel;k++){
11044: fscanf(ficpar,"%le",&delti3[i][j][k]);
11045: printf(" %le",delti3[i][j][k]);
11046: fprintf(ficparo," %le",delti3[i][j][k]);
11047: fprintf(ficlog," %le",delti3[i][j][k]);
11048: }
11049: fscanf(ficpar,"\n");
11050: numlinepar++;
11051: printf("\n");
11052: fprintf(ficparo,"\n");
11053: fprintf(ficlog,"\n");
1.126 brouard 11054: }
11055: }
11056: fflush(ficlog);
1.234 brouard 11057:
1.145 brouard 11058: /* Reads covariance matrix */
1.126 brouard 11059: delti=delti3[1][1];
1.220 brouard 11060:
11061:
1.126 brouard 11062: /* 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 11063:
1.126 brouard 11064: /* Reads comments: lines beginning with '#' */
11065: while((c=getc(ficpar))=='#' && c!= EOF){
11066: ungetc(c,ficpar);
11067: fgets(line, MAXLINE, ficpar);
11068: numlinepar++;
1.141 brouard 11069: fputs(line,stdout);
1.126 brouard 11070: fputs(line,ficparo);
11071: fputs(line,ficlog);
11072: }
11073: ungetc(c,ficpar);
1.220 brouard 11074:
1.126 brouard 11075: matcov=matrix(1,npar,1,npar);
1.203 brouard 11076: hess=matrix(1,npar,1,npar);
1.131 brouard 11077: for(i=1; i <=npar; i++)
11078: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11079:
1.194 brouard 11080: /* Scans npar lines */
1.126 brouard 11081: for(i=1; i <=npar; i++){
1.226 brouard 11082: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11083: if(count != 3){
1.226 brouard 11084: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11085: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11086: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11087: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11088: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11089: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11090: exit(1);
1.220 brouard 11091: }else{
1.226 brouard 11092: if(mle==1)
11093: printf("%1d%1d%d",i1,j1,jk);
11094: }
11095: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11096: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11097: for(j=1; j <=i; j++){
1.226 brouard 11098: fscanf(ficpar," %le",&matcov[i][j]);
11099: if(mle==1){
11100: printf(" %.5le",matcov[i][j]);
11101: }
11102: fprintf(ficlog," %.5le",matcov[i][j]);
11103: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11104: }
11105: fscanf(ficpar,"\n");
11106: numlinepar++;
11107: if(mle==1)
1.220 brouard 11108: printf("\n");
1.126 brouard 11109: fprintf(ficlog,"\n");
11110: fprintf(ficparo,"\n");
11111: }
1.194 brouard 11112: /* End of read covariance matrix npar lines */
1.126 brouard 11113: for(i=1; i <=npar; i++)
11114: for(j=i+1;j<=npar;j++)
1.226 brouard 11115: matcov[i][j]=matcov[j][i];
1.126 brouard 11116:
11117: if(mle==1)
11118: printf("\n");
11119: fprintf(ficlog,"\n");
11120:
11121: fflush(ficlog);
11122:
11123: /*-------- Rewriting parameter file ----------*/
11124: strcpy(rfileres,"r"); /* "Rparameterfile */
11125: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
11126: strcat(rfileres,"."); /* */
11127: strcat(rfileres,optionfilext); /* Other files have txt extension */
11128: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 11129: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11130: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 11131: }
11132: fprintf(ficres,"#%s\n",version);
11133: } /* End of mle != -3 */
1.218 brouard 11134:
1.186 brouard 11135: /* Main data
11136: */
1.126 brouard 11137: n= lastobs;
11138: num=lvector(1,n);
11139: moisnais=vector(1,n);
11140: annais=vector(1,n);
11141: moisdc=vector(1,n);
11142: andc=vector(1,n);
1.220 brouard 11143: weight=vector(1,n);
1.126 brouard 11144: agedc=vector(1,n);
11145: cod=ivector(1,n);
1.220 brouard 11146: for(i=1;i<=n;i++){
1.234 brouard 11147: num[i]=0;
11148: moisnais[i]=0;
11149: annais[i]=0;
11150: moisdc[i]=0;
11151: andc[i]=0;
11152: agedc[i]=0;
11153: cod[i]=0;
11154: weight[i]=1.0; /* Equal weights, 1 by default */
11155: }
1.126 brouard 11156: mint=matrix(1,maxwav,1,n);
11157: anint=matrix(1,maxwav,1,n);
1.131 brouard 11158: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11159: tab=ivector(1,NCOVMAX);
1.144 brouard 11160: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11161: 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 11162:
1.136 brouard 11163: /* Reads data from file datafile */
11164: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11165: goto end;
11166:
11167: /* Calculation of the number of parameters from char model */
1.234 brouard 11168: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11169: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11170: k=3 V4 Tvar[k=3]= 4 (from V4)
11171: k=2 V1 Tvar[k=2]= 1 (from V1)
11172: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11173: */
11174:
11175: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11176: TvarsDind=ivector(1,NCOVMAX); /* */
11177: TvarsD=ivector(1,NCOVMAX); /* */
11178: TvarsQind=ivector(1,NCOVMAX); /* */
11179: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11180: TvarF=ivector(1,NCOVMAX); /* */
11181: TvarFind=ivector(1,NCOVMAX); /* */
11182: TvarV=ivector(1,NCOVMAX); /* */
11183: TvarVind=ivector(1,NCOVMAX); /* */
11184: TvarA=ivector(1,NCOVMAX); /* */
11185: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11186: TvarFD=ivector(1,NCOVMAX); /* */
11187: TvarFDind=ivector(1,NCOVMAX); /* */
11188: TvarFQ=ivector(1,NCOVMAX); /* */
11189: TvarFQind=ivector(1,NCOVMAX); /* */
11190: TvarVD=ivector(1,NCOVMAX); /* */
11191: TvarVDind=ivector(1,NCOVMAX); /* */
11192: TvarVQ=ivector(1,NCOVMAX); /* */
11193: TvarVQind=ivector(1,NCOVMAX); /* */
11194:
1.230 brouard 11195: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11196: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11197: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11198: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11199: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11200: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11201: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11202: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11203: */
11204: /* For model-covariate k tells which data-covariate to use but
11205: because this model-covariate is a construction we invent a new column
11206: ncovcol + k1
11207: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11208: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11209: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11210: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11211: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11212: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11213: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11214: */
1.145 brouard 11215: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11216: 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 11217: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11218: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11219: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11220: 4 covariates (3 plus signs)
11221: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11222: */
1.230 brouard 11223: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11224: * individual dummy, fixed or varying:
11225: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11226: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11227: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11228: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11229: * Tmodelind[1]@9={9,0,3,2,}*/
11230: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11231: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11232: * individual quantitative, fixed or varying:
11233: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11234: * 3, 1, 0, 0, 0, 0, 0, 0},
11235: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11236: /* Main decodemodel */
11237:
1.187 brouard 11238:
1.223 brouard 11239: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11240: goto end;
11241:
1.137 brouard 11242: if((double)(lastobs-imx)/(double)imx > 1.10){
11243: nbwarn++;
11244: 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);
11245: 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);
11246: }
1.136 brouard 11247: /* if(mle==1){*/
1.137 brouard 11248: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11249: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11250: }
11251:
11252: /*-calculation of age at interview from date of interview and age at death -*/
11253: agev=matrix(1,maxwav,1,imx);
11254:
11255: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11256: goto end;
11257:
1.126 brouard 11258:
1.136 brouard 11259: agegomp=(int)agemin;
11260: free_vector(moisnais,1,n);
11261: free_vector(annais,1,n);
1.126 brouard 11262: /* free_matrix(mint,1,maxwav,1,n);
11263: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11264: /* free_vector(moisdc,1,n); */
11265: /* free_vector(andc,1,n); */
1.145 brouard 11266: /* */
11267:
1.126 brouard 11268: wav=ivector(1,imx);
1.214 brouard 11269: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11270: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11271: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11272: 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.*/
11273: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11274: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11275:
11276: /* Concatenates waves */
1.214 brouard 11277: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11278: Death is a valid wave (if date is known).
11279: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11280: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11281: and mw[mi+1][i]. dh depends on stepm.
11282: */
11283:
1.126 brouard 11284: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11285: /* Concatenates waves */
1.145 brouard 11286:
1.215 brouard 11287: free_vector(moisdc,1,n);
11288: free_vector(andc,1,n);
11289:
1.126 brouard 11290: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11291: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11292: ncodemax[1]=1;
1.145 brouard 11293: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11294: cptcoveff=0;
1.220 brouard 11295: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11296: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11297: }
11298:
11299: ncovcombmax=pow(2,cptcoveff);
11300: invalidvarcomb=ivector(1, ncovcombmax);
11301: for(i=1;i<ncovcombmax;i++)
11302: invalidvarcomb[i]=0;
11303:
1.211 brouard 11304: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11305: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11306: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11307:
1.200 brouard 11308: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11309: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11310: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11311: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11312: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11313: * (currently 0 or 1) in the data.
11314: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11315: * corresponding modality (h,j).
11316: */
11317:
1.145 brouard 11318: h=0;
11319: /*if (cptcovn > 0) */
1.126 brouard 11320: m=pow(2,cptcoveff);
11321:
1.144 brouard 11322: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11323: * For k=4 covariates, h goes from 1 to m=2**k
11324: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11325: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11326: * h\k 1 2 3 4
1.143 brouard 11327: *______________________________
11328: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11329: * 2 2 1 1 1
11330: * 3 i=2 1 2 1 1
11331: * 4 2 2 1 1
11332: * 5 i=3 1 i=2 1 2 1
11333: * 6 2 1 2 1
11334: * 7 i=4 1 2 2 1
11335: * 8 2 2 2 1
1.197 brouard 11336: * 9 i=5 1 i=3 1 i=2 1 2
11337: * 10 2 1 1 2
11338: * 11 i=6 1 2 1 2
11339: * 12 2 2 1 2
11340: * 13 i=7 1 i=4 1 2 2
11341: * 14 2 1 2 2
11342: * 15 i=8 1 2 2 2
11343: * 16 2 2 2 2
1.143 brouard 11344: */
1.212 brouard 11345: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11346: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11347: * and the value of each covariate?
11348: * V1=1, V2=1, V3=2, V4=1 ?
11349: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11350: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11351: * In order to get the real value in the data, we use nbcode
11352: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11353: * We are keeping this crazy system in order to be able (in the future?)
11354: * to have more than 2 values (0 or 1) for a covariate.
11355: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11356: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11357: * bbbbbbbb
11358: * 76543210
11359: * h-1 00000101 (6-1=5)
1.219 brouard 11360: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11361: * &
11362: * 1 00000001 (1)
1.219 brouard 11363: * 00000000 = 1 & ((h-1) >> (k-1))
11364: * +1= 00000001 =1
1.211 brouard 11365: *
11366: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11367: * h' 1101 =2^3+2^2+0x2^1+2^0
11368: * >>k' 11
11369: * & 00000001
11370: * = 00000001
11371: * +1 = 00000010=2 = codtabm(14,3)
11372: * Reverse h=6 and m=16?
11373: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11374: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11375: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11376: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11377: * V3=decodtabm(14,3,2**4)=2
11378: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11379: *(h-1) >> (j-1) 0011 =13 >> 2
11380: * &1 000000001
11381: * = 000000001
11382: * +1= 000000010 =2
11383: * 2211
11384: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11385: * V3=2
1.220 brouard 11386: * codtabm and decodtabm are identical
1.211 brouard 11387: */
11388:
1.145 brouard 11389:
11390: free_ivector(Ndum,-1,NCOVMAX);
11391:
11392:
1.126 brouard 11393:
1.186 brouard 11394: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11395: strcpy(optionfilegnuplot,optionfilefiname);
11396: if(mle==-3)
1.201 brouard 11397: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11398: strcat(optionfilegnuplot,".gp");
11399:
11400: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11401: printf("Problem with file %s",optionfilegnuplot);
11402: }
11403: else{
1.204 brouard 11404: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11405: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11406: //fprintf(ficgp,"set missing 'NaNq'\n");
11407: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11408: }
11409: /* fclose(ficgp);*/
1.186 brouard 11410:
11411:
11412: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11413:
11414: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11415: if(mle==-3)
1.201 brouard 11416: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11417: strcat(optionfilehtm,".htm");
11418: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11419: printf("Problem with %s \n",optionfilehtm);
11420: exit(0);
1.126 brouard 11421: }
11422:
11423: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11424: strcat(optionfilehtmcov,"-cov.htm");
11425: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11426: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11427: }
11428: else{
11429: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11430: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11431: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11432: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11433: }
11434:
1.213 brouard 11435: 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 11436: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11437: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11438: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11439: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11440: \n\
11441: <hr size=\"2\" color=\"#EC5E5E\">\
11442: <ul><li><h4>Parameter files</h4>\n\
11443: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11444: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11445: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11446: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11447: - Date and time at start: %s</ul>\n",\
11448: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11449: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11450: fileres,fileres,\
11451: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11452: fflush(fichtm);
11453:
11454: strcpy(pathr,path);
11455: strcat(pathr,optionfilefiname);
1.184 brouard 11456: #ifdef WIN32
11457: _chdir(optionfilefiname); /* Move to directory named optionfile */
11458: #else
1.126 brouard 11459: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11460: #endif
11461:
1.126 brouard 11462:
1.220 brouard 11463: /* Calculates basic frequencies. Computes observed prevalence at single age
11464: and for any valid combination of covariates
1.126 brouard 11465: and prints on file fileres'p'. */
1.251 brouard 11466: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11467: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11468:
11469: fprintf(fichtm,"\n");
11470: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
11471: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11472: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
11473: imx,agemin,agemax,jmin,jmax,jmean);
11474: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11475: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11476: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11477: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11478: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11479:
1.126 brouard 11480: /* For Powell, parameters are in a vector p[] starting at p[1]
11481: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11482: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11483:
11484: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11485: /* For mortality only */
1.126 brouard 11486: if (mle==-3){
1.136 brouard 11487: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11488: for(i=1;i<=NDIM;i++)
11489: for(j=1;j<=NDIM;j++)
11490: ximort[i][j]=0.;
1.186 brouard 11491: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11492: cens=ivector(1,n);
11493: ageexmed=vector(1,n);
11494: agecens=vector(1,n);
11495: dcwave=ivector(1,n);
1.223 brouard 11496:
1.126 brouard 11497: for (i=1; i<=imx; i++){
11498: dcwave[i]=-1;
11499: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11500: if (s[m][i]>nlstate) {
11501: dcwave[i]=m;
11502: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11503: break;
11504: }
1.126 brouard 11505: }
1.226 brouard 11506:
1.126 brouard 11507: for (i=1; i<=imx; i++) {
11508: if (wav[i]>0){
1.226 brouard 11509: ageexmed[i]=agev[mw[1][i]][i];
11510: j=wav[i];
11511: agecens[i]=1.;
11512:
11513: if (ageexmed[i]> 1 && wav[i] > 0){
11514: agecens[i]=agev[mw[j][i]][i];
11515: cens[i]= 1;
11516: }else if (ageexmed[i]< 1)
11517: cens[i]= -1;
11518: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11519: cens[i]=0 ;
1.126 brouard 11520: }
11521: else cens[i]=-1;
11522: }
11523:
11524: for (i=1;i<=NDIM;i++) {
11525: for (j=1;j<=NDIM;j++)
1.226 brouard 11526: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11527: }
11528:
1.145 brouard 11529: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11530: /*printf("%lf %lf", p[1], p[2]);*/
11531:
11532:
1.136 brouard 11533: #ifdef GSL
11534: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11535: #else
1.126 brouard 11536: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11537: #endif
1.201 brouard 11538: strcpy(filerespow,"POW-MORT_");
11539: strcat(filerespow,fileresu);
1.126 brouard 11540: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11541: printf("Problem with resultfile: %s\n", filerespow);
11542: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11543: }
1.136 brouard 11544: #ifdef GSL
11545: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11546: #else
1.126 brouard 11547: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11548: #endif
1.126 brouard 11549: /* for (i=1;i<=nlstate;i++)
11550: for(j=1;j<=nlstate+ndeath;j++)
11551: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11552: */
11553: fprintf(ficrespow,"\n");
1.136 brouard 11554: #ifdef GSL
11555: /* gsl starts here */
11556: T = gsl_multimin_fminimizer_nmsimplex;
11557: gsl_multimin_fminimizer *sfm = NULL;
11558: gsl_vector *ss, *x;
11559: gsl_multimin_function minex_func;
11560:
11561: /* Initial vertex size vector */
11562: ss = gsl_vector_alloc (NDIM);
11563:
11564: if (ss == NULL){
11565: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11566: }
11567: /* Set all step sizes to 1 */
11568: gsl_vector_set_all (ss, 0.001);
11569:
11570: /* Starting point */
1.126 brouard 11571:
1.136 brouard 11572: x = gsl_vector_alloc (NDIM);
11573:
11574: if (x == NULL){
11575: gsl_vector_free(ss);
11576: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11577: }
11578:
11579: /* Initialize method and iterate */
11580: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11581: /* gsl_vector_set(x, 0, 0.0268); */
11582: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11583: gsl_vector_set(x, 0, p[1]);
11584: gsl_vector_set(x, 1, p[2]);
11585:
11586: minex_func.f = &gompertz_f;
11587: minex_func.n = NDIM;
11588: minex_func.params = (void *)&p; /* ??? */
11589:
11590: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11591: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11592:
11593: printf("Iterations beginning .....\n\n");
11594: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11595:
11596: iteri=0;
11597: while (rval == GSL_CONTINUE){
11598: iteri++;
11599: status = gsl_multimin_fminimizer_iterate(sfm);
11600:
11601: if (status) printf("error: %s\n", gsl_strerror (status));
11602: fflush(0);
11603:
11604: if (status)
11605: break;
11606:
11607: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11608: ssval = gsl_multimin_fminimizer_size (sfm);
11609:
11610: if (rval == GSL_SUCCESS)
11611: printf ("converged to a local maximum at\n");
11612:
11613: printf("%5d ", iteri);
11614: for (it = 0; it < NDIM; it++){
11615: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11616: }
11617: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11618: }
11619:
11620: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11621:
11622: gsl_vector_free(x); /* initial values */
11623: gsl_vector_free(ss); /* inital step size */
11624: for (it=0; it<NDIM; it++){
11625: p[it+1]=gsl_vector_get(sfm->x,it);
11626: fprintf(ficrespow," %.12lf", p[it]);
11627: }
11628: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11629: #endif
11630: #ifdef POWELL
11631: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11632: #endif
1.126 brouard 11633: fclose(ficrespow);
11634:
1.203 brouard 11635: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11636:
11637: for(i=1; i <=NDIM; i++)
11638: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11639: matcov[i][j]=matcov[j][i];
1.126 brouard 11640:
11641: printf("\nCovariance matrix\n ");
1.203 brouard 11642: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11643: for(i=1; i <=NDIM; i++) {
11644: for(j=1;j<=NDIM;j++){
1.220 brouard 11645: printf("%f ",matcov[i][j]);
11646: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11647: }
1.203 brouard 11648: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11649: }
11650:
11651: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11652: for (i=1;i<=NDIM;i++) {
1.126 brouard 11653: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11654: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11655: }
1.126 brouard 11656: lsurv=vector(1,AGESUP);
11657: lpop=vector(1,AGESUP);
11658: tpop=vector(1,AGESUP);
11659: lsurv[agegomp]=100000;
11660:
11661: for (k=agegomp;k<=AGESUP;k++) {
11662: agemortsup=k;
11663: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11664: }
11665:
11666: for (k=agegomp;k<agemortsup;k++)
11667: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11668:
11669: for (k=agegomp;k<agemortsup;k++){
11670: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11671: sumlpop=sumlpop+lpop[k];
11672: }
11673:
11674: tpop[agegomp]=sumlpop;
11675: for (k=agegomp;k<(agemortsup-3);k++){
11676: /* tpop[k+1]=2;*/
11677: tpop[k+1]=tpop[k]-lpop[k];
11678: }
11679:
11680:
11681: printf("\nAge lx qx dx Lx Tx e(x)\n");
11682: for (k=agegomp;k<(agemortsup-2);k++)
11683: 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]);
11684:
11685:
11686: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11687: ageminpar=50;
11688: agemaxpar=100;
1.194 brouard 11689: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11690: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11691: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11692: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11693: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11694: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11695: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11696: }else{
11697: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11698: 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 11699: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11700: }
1.201 brouard 11701: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11702: stepm, weightopt,\
11703: model,imx,p,matcov,agemortsup);
11704:
11705: free_vector(lsurv,1,AGESUP);
11706: free_vector(lpop,1,AGESUP);
11707: free_vector(tpop,1,AGESUP);
1.220 brouard 11708: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11709: free_ivector(cens,1,n);
11710: free_vector(agecens,1,n);
11711: free_ivector(dcwave,1,n);
1.220 brouard 11712: #ifdef GSL
1.136 brouard 11713: #endif
1.186 brouard 11714: } /* Endof if mle==-3 mortality only */
1.205 brouard 11715: /* Standard */
11716: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11717: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11718: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11719: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11720: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11721: for (k=1; k<=npar;k++)
11722: printf(" %d %8.5f",k,p[k]);
11723: printf("\n");
1.205 brouard 11724: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11725: /* mlikeli uses func not funcone */
1.247 brouard 11726: /* for(i=1;i<nlstate;i++){ */
11727: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11728: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11729: /* } */
1.205 brouard 11730: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11731: }
11732: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11733: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11734: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11735: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11736: }
11737: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11738: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11739: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11740: for (k=1; k<=npar;k++)
11741: printf(" %d %8.5f",k,p[k]);
11742: printf("\n");
11743:
11744: /*--------- results files --------------*/
1.224 brouard 11745: 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 11746:
11747:
11748: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11749: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11750: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11751: for(i=1,jk=1; i <=nlstate; i++){
11752: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11753: if (k != i) {
11754: printf("%d%d ",i,k);
11755: fprintf(ficlog,"%d%d ",i,k);
11756: fprintf(ficres,"%1d%1d ",i,k);
11757: for(j=1; j <=ncovmodel; j++){
11758: printf("%12.7f ",p[jk]);
11759: fprintf(ficlog,"%12.7f ",p[jk]);
11760: fprintf(ficres,"%12.7f ",p[jk]);
11761: jk++;
11762: }
11763: printf("\n");
11764: fprintf(ficlog,"\n");
11765: fprintf(ficres,"\n");
11766: }
1.126 brouard 11767: }
11768: }
1.203 brouard 11769: if(mle != 0){
11770: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11771: ftolhess=ftol; /* Usually correct */
1.203 brouard 11772: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11773: 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");
11774: 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");
11775: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11776: for(k=1; k <=(nlstate+ndeath); k++){
11777: if (k != i) {
11778: printf("%d%d ",i,k);
11779: fprintf(ficlog,"%d%d ",i,k);
11780: for(j=1; j <=ncovmodel; j++){
11781: 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]));
11782: 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]));
11783: jk++;
11784: }
11785: printf("\n");
11786: fprintf(ficlog,"\n");
11787: }
11788: }
1.193 brouard 11789: }
1.203 brouard 11790: } /* end of hesscov and Wald tests */
1.225 brouard 11791:
1.203 brouard 11792: /* */
1.126 brouard 11793: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11794: printf("# Scales (for hessian or gradient estimation)\n");
11795: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11796: for(i=1,jk=1; i <=nlstate; i++){
11797: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11798: if (j!=i) {
11799: fprintf(ficres,"%1d%1d",i,j);
11800: printf("%1d%1d",i,j);
11801: fprintf(ficlog,"%1d%1d",i,j);
11802: for(k=1; k<=ncovmodel;k++){
11803: printf(" %.5e",delti[jk]);
11804: fprintf(ficlog," %.5e",delti[jk]);
11805: fprintf(ficres," %.5e",delti[jk]);
11806: jk++;
11807: }
11808: printf("\n");
11809: fprintf(ficlog,"\n");
11810: fprintf(ficres,"\n");
11811: }
1.126 brouard 11812: }
11813: }
11814:
11815: 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 11816: if(mle >= 1) /* To big for the screen */
1.126 brouard 11817: 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");
11818: 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");
11819: /* # 121 Var(a12)\n\ */
11820: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11821: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11822: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11823: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11824: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11825: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11826: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11827:
11828:
11829: /* Just to have a covariance matrix which will be more understandable
11830: even is we still don't want to manage dictionary of variables
11831: */
11832: for(itimes=1;itimes<=2;itimes++){
11833: jj=0;
11834: for(i=1; i <=nlstate; i++){
1.225 brouard 11835: for(j=1; j <=nlstate+ndeath; j++){
11836: if(j==i) continue;
11837: for(k=1; k<=ncovmodel;k++){
11838: jj++;
11839: ca[0]= k+'a'-1;ca[1]='\0';
11840: if(itimes==1){
11841: if(mle>=1)
11842: printf("#%1d%1d%d",i,j,k);
11843: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11844: fprintf(ficres,"#%1d%1d%d",i,j,k);
11845: }else{
11846: if(mle>=1)
11847: printf("%1d%1d%d",i,j,k);
11848: fprintf(ficlog,"%1d%1d%d",i,j,k);
11849: fprintf(ficres,"%1d%1d%d",i,j,k);
11850: }
11851: ll=0;
11852: for(li=1;li <=nlstate; li++){
11853: for(lj=1;lj <=nlstate+ndeath; lj++){
11854: if(lj==li) continue;
11855: for(lk=1;lk<=ncovmodel;lk++){
11856: ll++;
11857: if(ll<=jj){
11858: cb[0]= lk +'a'-1;cb[1]='\0';
11859: if(ll<jj){
11860: if(itimes==1){
11861: if(mle>=1)
11862: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11863: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11864: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11865: }else{
11866: if(mle>=1)
11867: printf(" %.5e",matcov[jj][ll]);
11868: fprintf(ficlog," %.5e",matcov[jj][ll]);
11869: fprintf(ficres," %.5e",matcov[jj][ll]);
11870: }
11871: }else{
11872: if(itimes==1){
11873: if(mle>=1)
11874: printf(" Var(%s%1d%1d)",ca,i,j);
11875: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11876: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11877: }else{
11878: if(mle>=1)
11879: printf(" %.7e",matcov[jj][ll]);
11880: fprintf(ficlog," %.7e",matcov[jj][ll]);
11881: fprintf(ficres," %.7e",matcov[jj][ll]);
11882: }
11883: }
11884: }
11885: } /* end lk */
11886: } /* end lj */
11887: } /* end li */
11888: if(mle>=1)
11889: printf("\n");
11890: fprintf(ficlog,"\n");
11891: fprintf(ficres,"\n");
11892: numlinepar++;
11893: } /* end k*/
11894: } /*end j */
1.126 brouard 11895: } /* end i */
11896: } /* end itimes */
11897:
11898: fflush(ficlog);
11899: fflush(ficres);
1.225 brouard 11900: while(fgets(line, MAXLINE, ficpar)) {
11901: /* If line starts with a # it is a comment */
11902: if (line[0] == '#') {
11903: numlinepar++;
11904: fputs(line,stdout);
11905: fputs(line,ficparo);
11906: fputs(line,ficlog);
11907: continue;
11908: }else
11909: break;
11910: }
11911:
1.209 brouard 11912: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11913: /* ungetc(c,ficpar); */
11914: /* fgets(line, MAXLINE, ficpar); */
11915: /* fputs(line,stdout); */
11916: /* fputs(line,ficparo); */
11917: /* } */
11918: /* ungetc(c,ficpar); */
1.126 brouard 11919:
11920: estepm=0;
1.209 brouard 11921: 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 11922:
11923: if (num_filled != 6) {
11924: 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);
11925: 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);
11926: goto end;
11927: }
11928: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11929: }
11930: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11931: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11932:
1.209 brouard 11933: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11934: if (estepm==0 || estepm < stepm) estepm=stepm;
11935: if (fage <= 2) {
11936: bage = ageminpar;
11937: fage = agemaxpar;
11938: }
11939:
11940: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11941: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11942: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11943:
1.186 brouard 11944: /* Other stuffs, more or less useful */
1.254 brouard 11945: while(fgets(line, MAXLINE, ficpar)) {
11946: /* If line starts with a # it is a comment */
11947: if (line[0] == '#') {
11948: numlinepar++;
11949: fputs(line,stdout);
11950: fputs(line,ficparo);
11951: fputs(line,ficlog);
11952: continue;
11953: }else
11954: break;
11955: }
11956:
11957: 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){
11958:
11959: if (num_filled != 7) {
11960: 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);
11961: 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);
11962: goto end;
11963: }
11964: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11965: 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);
11966: 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);
11967: 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 11968: }
1.254 brouard 11969:
11970: while(fgets(line, MAXLINE, ficpar)) {
11971: /* If line starts with a # it is a comment */
11972: if (line[0] == '#') {
11973: numlinepar++;
11974: fputs(line,stdout);
11975: fputs(line,ficparo);
11976: fputs(line,ficlog);
11977: continue;
11978: }else
11979: break;
1.126 brouard 11980: }
11981:
11982:
11983: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11984: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11985:
1.254 brouard 11986: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11987: if (num_filled != 1) {
11988: 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);
11989: 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);
11990: goto end;
11991: }
11992: printf("pop_based=%d\n",popbased);
11993: fprintf(ficlog,"pop_based=%d\n",popbased);
11994: fprintf(ficparo,"pop_based=%d\n",popbased);
11995: fprintf(ficres,"pop_based=%d\n",popbased);
11996: }
11997:
1.258 brouard 11998: /* Results */
11999: nresult=0;
12000: do{
12001: if(!fgets(line, MAXLINE, ficpar)){
12002: endishere=1;
12003: parameterline=14;
12004: }else if (line[0] == '#') {
12005: /* If line starts with a # it is a comment */
1.254 brouard 12006: numlinepar++;
12007: fputs(line,stdout);
12008: fputs(line,ficparo);
12009: fputs(line,ficlog);
12010: continue;
1.258 brouard 12011: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12012: parameterline=11;
12013: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12014: parameterline=12;
12015: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12016: parameterline=13;
12017: else{
12018: parameterline=14;
1.254 brouard 12019: }
1.258 brouard 12020: switch (parameterline){
12021: case 11:
12022: 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){
12023: if (num_filled != 8) {
12024: 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);
12025: 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);
12026: goto end;
12027: }
12028: 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);
12029: 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);
12030: 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);
12031: 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);
12032: /* day and month of proj2 are not used but only year anproj2.*/
1.273 ! brouard 12033: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
! 12034: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
! 12035:
1.258 brouard 12036: }
1.254 brouard 12037: break;
1.258 brouard 12038: case 12:
12039: /*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);*/
12040: 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){
12041: if (num_filled != 8) {
1.262 brouard 12042: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12043: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
1.258 brouard 12044: goto end;
12045: }
12046: 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);
12047: 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);
12048: 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);
12049: 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);
12050: /* day and month of proj2 are not used but only year anproj2.*/
1.273 ! brouard 12051: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
! 12052: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12053: }
1.230 brouard 12054: break;
1.258 brouard 12055: case 13:
12056: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12057: if (num_filled == 0){
12058: resultline[0]='\0';
12059: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12060: 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);
12061: break;
12062: } else if (num_filled != 1){
12063: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12064: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12065: }
12066: nresult++; /* Sum of resultlines */
12067: printf("Result %d: result=%s\n",nresult, resultline);
12068: if(nresult > MAXRESULTLINES){
12069: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12070: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12071: goto end;
12072: }
12073: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12074: fprintf(ficparo,"result: %s\n",resultline);
12075: fprintf(ficres,"result: %s\n",resultline);
12076: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12077: break;
1.258 brouard 12078: case 14:
1.259 brouard 12079: if(ncovmodel >2 && nresult==0 ){
12080: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12081: goto end;
12082: }
1.259 brouard 12083: break;
1.258 brouard 12084: default:
12085: nresult=1;
12086: decoderesult(".",nresult ); /* No covariate */
12087: }
12088: } /* End switch parameterline */
12089: }while(endishere==0); /* End do */
1.126 brouard 12090:
1.230 brouard 12091: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12092: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12093:
12094: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12095: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12096: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12097: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12098: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12099: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12100: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12101: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12102: }else{
1.270 brouard 12103: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12104: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12105: }
12106: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12107: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 ! brouard 12108: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12109:
1.225 brouard 12110: /*------------ free_vector -------------*/
12111: /* chdir(path); */
1.220 brouard 12112:
1.215 brouard 12113: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12114: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12115: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12116: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12117: free_lvector(num,1,n);
12118: free_vector(agedc,1,n);
12119: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12120: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12121: fclose(ficparo);
12122: fclose(ficres);
1.220 brouard 12123:
12124:
1.186 brouard 12125: /* Other results (useful)*/
1.220 brouard 12126:
12127:
1.126 brouard 12128: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12129: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12130: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12131: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12132: fclose(ficrespl);
12133:
12134: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12135: /*#include "hpijx.h"*/
12136: hPijx(p, bage, fage);
1.145 brouard 12137: fclose(ficrespij);
1.227 brouard 12138:
1.220 brouard 12139: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12140: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12141: k=1;
1.126 brouard 12142: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12143:
1.269 brouard 12144: /* Prevalence for each covariate combination in probs[age][status][cov] */
12145: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12146: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12147: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12148: for(k=1;k<=ncovcombmax;k++)
12149: probs[i][j][k]=0.;
1.269 brouard 12150: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12151: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12152: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12153: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12154: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12155: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12156: for(k=1;k<=ncovcombmax;k++)
12157: mobaverages[i][j][k]=0.;
1.219 brouard 12158: mobaverage=mobaverages;
12159: if (mobilav!=0) {
1.235 brouard 12160: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12161: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12162: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12163: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12164: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12165: }
1.269 brouard 12166: } else if (mobilavproj !=0) {
1.235 brouard 12167: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12168: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12169: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12170: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12171: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12172: }
1.269 brouard 12173: }else{
12174: printf("Internal error moving average\n");
12175: fflush(stdout);
12176: exit(1);
1.219 brouard 12177: }
12178: }/* end if moving average */
1.227 brouard 12179:
1.126 brouard 12180: /*---------- Forecasting ------------------*/
12181: if(prevfcast==1){
12182: /* if(stepm ==1){*/
1.269 brouard 12183: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12184: }
1.269 brouard 12185:
12186: /* Backcasting */
1.217 brouard 12187: if(backcast==1){
1.219 brouard 12188: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12189: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12190: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12191:
12192: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12193:
12194: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12195:
1.219 brouard 12196: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12197: fclose(ficresplb);
12198:
1.222 brouard 12199: hBijx(p, bage, fage, mobaverage);
12200: fclose(ficrespijb);
1.219 brouard 12201:
1.269 brouard 12202: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12203: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12204: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12205:
12206:
1.269 brouard 12207: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12208: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12209: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12210: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12211: } /* end Backcasting */
1.268 brouard 12212:
1.186 brouard 12213:
12214: /* ------ Other prevalence ratios------------ */
1.126 brouard 12215:
1.215 brouard 12216: free_ivector(wav,1,imx);
12217: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12218: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12219: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12220:
12221:
1.127 brouard 12222: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12223:
1.201 brouard 12224: strcpy(filerese,"E_");
12225: strcat(filerese,fileresu);
1.126 brouard 12226: if((ficreseij=fopen(filerese,"w"))==NULL) {
12227: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12228: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12229: }
1.208 brouard 12230: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12231: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12232:
12233: pstamp(ficreseij);
1.219 brouard 12234:
1.235 brouard 12235: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12236: if (cptcovn < 1){i1=1;}
12237:
12238: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12239: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12240: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12241: continue;
1.219 brouard 12242: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12243: printf("\n#****** ");
1.225 brouard 12244: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12245: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12246: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12247: }
12248: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12249: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12250: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12251: }
12252: fprintf(ficreseij,"******\n");
1.235 brouard 12253: printf("******\n");
1.219 brouard 12254:
12255: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12256: oldm=oldms;savm=savms;
1.235 brouard 12257: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12258:
1.219 brouard 12259: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12260: }
12261: fclose(ficreseij);
1.208 brouard 12262: printf("done evsij\n");fflush(stdout);
12263: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12264:
1.218 brouard 12265:
1.227 brouard 12266: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12267:
1.201 brouard 12268: strcpy(filerest,"T_");
12269: strcat(filerest,fileresu);
1.127 brouard 12270: if((ficrest=fopen(filerest,"w"))==NULL) {
12271: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12272: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12273: }
1.208 brouard 12274: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12275: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12276: strcpy(fileresstde,"STDE_");
12277: strcat(fileresstde,fileresu);
1.126 brouard 12278: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12279: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12280: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12281: }
1.227 brouard 12282: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12283: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12284:
1.201 brouard 12285: strcpy(filerescve,"CVE_");
12286: strcat(filerescve,fileresu);
1.126 brouard 12287: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12288: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12289: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12290: }
1.227 brouard 12291: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12292: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12293:
1.201 brouard 12294: strcpy(fileresv,"V_");
12295: strcat(fileresv,fileresu);
1.126 brouard 12296: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12297: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12298: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12299: }
1.227 brouard 12300: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12301: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12302:
1.235 brouard 12303: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12304: if (cptcovn < 1){i1=1;}
12305:
12306: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12307: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12308: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12309: continue;
1.242 brouard 12310: printf("\n#****** Result for:");
12311: fprintf(ficrest,"\n#****** Result for:");
12312: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12313: for(j=1;j<=cptcoveff;j++){
12314: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12315: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12316: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12317: }
1.235 brouard 12318: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12319: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12320: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12321: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12322: }
1.208 brouard 12323: fprintf(ficrest,"******\n");
1.227 brouard 12324: fprintf(ficlog,"******\n");
12325: printf("******\n");
1.208 brouard 12326:
12327: fprintf(ficresstdeij,"\n#****** ");
12328: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12329: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12330: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12331: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12332: }
1.235 brouard 12333: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12334: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12335: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12336: }
1.208 brouard 12337: fprintf(ficresstdeij,"******\n");
12338: fprintf(ficrescveij,"******\n");
12339:
12340: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12341: /* pstamp(ficresvij); */
1.225 brouard 12342: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12343: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12344: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12345: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12346: }
1.208 brouard 12347: fprintf(ficresvij,"******\n");
12348:
12349: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12350: oldm=oldms;savm=savms;
1.235 brouard 12351: printf(" cvevsij ");
12352: fprintf(ficlog, " cvevsij ");
12353: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12354: printf(" end cvevsij \n ");
12355: fprintf(ficlog, " end cvevsij \n ");
12356:
12357: /*
12358: */
12359: /* goto endfree; */
12360:
12361: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12362: pstamp(ficrest);
12363:
1.269 brouard 12364: epj=vector(1,nlstate+1);
1.208 brouard 12365: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12366: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12367: cptcod= 0; /* To be deleted */
12368: printf("varevsij vpopbased=%d \n",vpopbased);
12369: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12370: 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 12371: 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 ");
12372: if(vpopbased==1)
12373: 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);
12374: else
12375: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12376: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12377: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12378: fprintf(ficrest,"\n");
12379: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12380: printf("Computing age specific period (stable) prevalences in each health state \n");
12381: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12382: for(age=bage; age <=fage ;age++){
1.235 brouard 12383: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12384: if (vpopbased==1) {
12385: if(mobilav ==0){
12386: for(i=1; i<=nlstate;i++)
12387: prlim[i][i]=probs[(int)age][i][k];
12388: }else{ /* mobilav */
12389: for(i=1; i<=nlstate;i++)
12390: prlim[i][i]=mobaverage[(int)age][i][k];
12391: }
12392: }
1.219 brouard 12393:
1.227 brouard 12394: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12395: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12396: /* printf(" age %4.0f ",age); */
12397: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12398: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12399: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12400: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12401: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12402: }
12403: epj[nlstate+1] +=epj[j];
12404: }
12405: /* printf(" age %4.0f \n",age); */
1.219 brouard 12406:
1.227 brouard 12407: for(i=1, vepp=0.;i <=nlstate;i++)
12408: for(j=1;j <=nlstate;j++)
12409: vepp += vareij[i][j][(int)age];
12410: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12411: for(j=1;j <=nlstate;j++){
12412: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12413: }
12414: fprintf(ficrest,"\n");
12415: }
1.208 brouard 12416: } /* End vpopbased */
1.269 brouard 12417: free_vector(epj,1,nlstate+1);
1.208 brouard 12418: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12419: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12420: printf("done selection\n");fflush(stdout);
12421: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12422:
1.235 brouard 12423: } /* End k selection */
1.227 brouard 12424:
12425: printf("done State-specific expectancies\n");fflush(stdout);
12426: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12427:
1.269 brouard 12428: /* variance-covariance of period prevalence*/
12429: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12430:
1.227 brouard 12431:
12432: free_vector(weight,1,n);
12433: free_imatrix(Tvard,1,NCOVMAX,1,2);
12434: free_imatrix(s,1,maxwav+1,1,n);
12435: free_matrix(anint,1,maxwav,1,n);
12436: free_matrix(mint,1,maxwav,1,n);
12437: free_ivector(cod,1,n);
12438: free_ivector(tab,1,NCOVMAX);
12439: fclose(ficresstdeij);
12440: fclose(ficrescveij);
12441: fclose(ficresvij);
12442: fclose(ficrest);
12443: fclose(ficpar);
12444:
12445:
1.126 brouard 12446: /*---------- End : free ----------------*/
1.219 brouard 12447: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12448: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12449: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12450: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12451: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12452: } /* mle==-3 arrives here for freeing */
1.227 brouard 12453: /* endfree:*/
12454: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12455: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12456: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12457: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12458: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12459: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12460: free_matrix(covar,0,NCOVMAX,1,n);
12461: free_matrix(matcov,1,npar,1,npar);
12462: free_matrix(hess,1,npar,1,npar);
12463: /*free_vector(delti,1,npar);*/
12464: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12465: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12466: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12467: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12468:
12469: free_ivector(ncodemax,1,NCOVMAX);
12470: free_ivector(ncodemaxwundef,1,NCOVMAX);
12471: free_ivector(Dummy,-1,NCOVMAX);
12472: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12473: free_ivector(DummyV,1,NCOVMAX);
12474: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12475: free_ivector(Typevar,-1,NCOVMAX);
12476: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12477: free_ivector(TvarsQ,1,NCOVMAX);
12478: free_ivector(TvarsQind,1,NCOVMAX);
12479: free_ivector(TvarsD,1,NCOVMAX);
12480: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12481: free_ivector(TvarFD,1,NCOVMAX);
12482: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12483: free_ivector(TvarF,1,NCOVMAX);
12484: free_ivector(TvarFind,1,NCOVMAX);
12485: free_ivector(TvarV,1,NCOVMAX);
12486: free_ivector(TvarVind,1,NCOVMAX);
12487: free_ivector(TvarA,1,NCOVMAX);
12488: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12489: free_ivector(TvarFQ,1,NCOVMAX);
12490: free_ivector(TvarFQind,1,NCOVMAX);
12491: free_ivector(TvarVD,1,NCOVMAX);
12492: free_ivector(TvarVDind,1,NCOVMAX);
12493: free_ivector(TvarVQ,1,NCOVMAX);
12494: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12495: free_ivector(Tvarsel,1,NCOVMAX);
12496: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12497: free_ivector(Tposprod,1,NCOVMAX);
12498: free_ivector(Tprod,1,NCOVMAX);
12499: free_ivector(Tvaraff,1,NCOVMAX);
12500: free_ivector(invalidvarcomb,1,ncovcombmax);
12501: free_ivector(Tage,1,NCOVMAX);
12502: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12503: free_ivector(TmodelInvind,1,NCOVMAX);
12504: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12505:
12506: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12507: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12508: fflush(fichtm);
12509: fflush(ficgp);
12510:
1.227 brouard 12511:
1.126 brouard 12512: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12513: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12514: 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 12515: }else{
12516: printf("End of Imach\n");
12517: fprintf(ficlog,"End of Imach\n");
12518: }
12519: printf("See log file on %s\n",filelog);
12520: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12521: /*(void) gettimeofday(&end_time,&tzp);*/
12522: rend_time = time(NULL);
12523: end_time = *localtime(&rend_time);
12524: /* tml = *localtime(&end_time.tm_sec); */
12525: strcpy(strtend,asctime(&end_time));
1.126 brouard 12526: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12527: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12528: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12529:
1.157 brouard 12530: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12531: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12532: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12533: /* printf("Total time was %d uSec.\n", total_usecs);*/
12534: /* if(fileappend(fichtm,optionfilehtm)){ */
12535: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12536: fclose(fichtm);
12537: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12538: fclose(fichtmcov);
12539: fclose(ficgp);
12540: fclose(ficlog);
12541: /*------ End -----------*/
1.227 brouard 12542:
12543:
12544: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12545: #ifdef WIN32
1.227 brouard 12546: if (_chdir(pathcd) != 0)
12547: printf("Can't move to directory %s!\n",path);
12548: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12549: #else
1.227 brouard 12550: if(chdir(pathcd) != 0)
12551: printf("Can't move to directory %s!\n", path);
12552: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12553: #endif
1.126 brouard 12554: printf("Current directory %s!\n",pathcd);
12555: /*strcat(plotcmd,CHARSEPARATOR);*/
12556: sprintf(plotcmd,"gnuplot");
1.157 brouard 12557: #ifdef _WIN32
1.126 brouard 12558: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12559: #endif
12560: if(!stat(plotcmd,&info)){
1.158 brouard 12561: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12562: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12563: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12564: }else
12565: strcpy(pplotcmd,plotcmd);
1.157 brouard 12566: #ifdef __unix
1.126 brouard 12567: strcpy(plotcmd,GNUPLOTPROGRAM);
12568: if(!stat(plotcmd,&info)){
1.158 brouard 12569: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12570: }else
12571: strcpy(pplotcmd,plotcmd);
12572: #endif
12573: }else
12574: strcpy(pplotcmd,plotcmd);
12575:
12576: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12577: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12578:
1.126 brouard 12579: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12580: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12581: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12582: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12583: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12584: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12585: }
1.158 brouard 12586: printf(" Successful, please wait...");
1.126 brouard 12587: while (z[0] != 'q') {
12588: /* chdir(path); */
1.154 brouard 12589: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12590: scanf("%s",z);
12591: /* if (z[0] == 'c') system("./imach"); */
12592: if (z[0] == 'e') {
1.158 brouard 12593: #ifdef __APPLE__
1.152 brouard 12594: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12595: #elif __linux
12596: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12597: #else
1.152 brouard 12598: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12599: #endif
12600: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12601: system(pplotcmd);
1.126 brouard 12602: }
12603: else if (z[0] == 'g') system(plotcmd);
12604: else if (z[0] == 'q') exit(0);
12605: }
1.227 brouard 12606: end:
1.126 brouard 12607: while (z[0] != 'q') {
1.195 brouard 12608: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12609: scanf("%s",z);
12610: }
12611: }
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