Annotation of imach/src/imach.c, revision 1.302
1.302 ! brouard 1: /* $Id: imach.c,v 1.301 2019/06/04 13:51:20 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.302 ! brouard 4: Revision 1.301 2019/06/04 13:51:20 brouard
! 5: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
! 6:
1.301 brouard 7: Revision 1.300 2019/05/22 19:09:45 brouard
8: Summary: version 0.99r19 of May 2019
9:
1.300 brouard 10: Revision 1.299 2019/05/22 18:37:08 brouard
11: Summary: Cleaned 0.99r19
12:
1.299 brouard 13: Revision 1.298 2019/05/22 18:19:56 brouard
14: *** empty log message ***
15:
1.298 brouard 16: Revision 1.297 2019/05/22 17:56:10 brouard
17: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
18:
1.297 brouard 19: Revision 1.296 2019/05/20 13:03:18 brouard
20: Summary: Projection syntax simplified
21:
22:
23: We can now start projections, forward or backward, from the mean date
24: of inteviews up to or down to a number of years of projection:
25: prevforecast=1 yearsfproj=15.3 mobil_average=0
26: or
27: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
28: or
29: prevbackcast=1 yearsbproj=12.3 mobil_average=1
30: or
31: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
32:
1.296 brouard 33: Revision 1.295 2019/05/18 09:52:50 brouard
34: Summary: doxygen tex bug
35:
1.295 brouard 36: Revision 1.294 2019/05/16 14:54:33 brouard
37: Summary: There was some wrong lines added
38:
1.294 brouard 39: Revision 1.293 2019/05/09 15:17:34 brouard
40: *** empty log message ***
41:
1.293 brouard 42: Revision 1.292 2019/05/09 14:17:20 brouard
43: Summary: Some updates
44:
1.292 brouard 45: Revision 1.291 2019/05/09 13:44:18 brouard
46: Summary: Before ncovmax
47:
1.291 brouard 48: Revision 1.290 2019/05/09 13:39:37 brouard
49: Summary: 0.99r18 unlimited number of individuals
50:
51: The number n which was limited to 20,000 cases is now unlimited, from firstobs to lastobs. If the number is too for the virtual memory, probably an error will occur.
52:
1.290 brouard 53: Revision 1.289 2018/12/13 09:16:26 brouard
54: Summary: Bug for young ages (<-30) will be in r17
55:
1.289 brouard 56: Revision 1.288 2018/05/02 20:58:27 brouard
57: Summary: Some bugs fixed
58:
1.288 brouard 59: Revision 1.287 2018/05/01 17:57:25 brouard
60: Summary: Bug fixed by providing frequencies only for non missing covariates
61:
1.287 brouard 62: Revision 1.286 2018/04/27 14:27:04 brouard
63: Summary: some minor bugs
64:
1.286 brouard 65: Revision 1.285 2018/04/21 21:02:16 brouard
66: Summary: Some bugs fixed, valgrind tested
67:
1.285 brouard 68: Revision 1.284 2018/04/20 05:22:13 brouard
69: Summary: Computing mean and stdeviation of fixed quantitative variables
70:
1.284 brouard 71: Revision 1.283 2018/04/19 14:49:16 brouard
72: Summary: Some minor bugs fixed
73:
1.283 brouard 74: Revision 1.282 2018/02/27 22:50:02 brouard
75: *** empty log message ***
76:
1.282 brouard 77: Revision 1.281 2018/02/27 19:25:23 brouard
78: Summary: Adding second argument for quitting
79:
1.281 brouard 80: Revision 1.280 2018/02/21 07:58:13 brouard
81: Summary: 0.99r15
82:
83: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
84:
1.280 brouard 85: Revision 1.279 2017/07/20 13:35:01 brouard
86: Summary: temporary working
87:
1.279 brouard 88: Revision 1.278 2017/07/19 14:09:02 brouard
89: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
90:
1.278 brouard 91: Revision 1.277 2017/07/17 08:53:49 brouard
92: Summary: BOM files can be read now
93:
1.277 brouard 94: Revision 1.276 2017/06/30 15:48:31 brouard
95: Summary: Graphs improvements
96:
1.276 brouard 97: Revision 1.275 2017/06/30 13:39:33 brouard
98: Summary: Saito's color
99:
1.275 brouard 100: Revision 1.274 2017/06/29 09:47:08 brouard
101: Summary: Version 0.99r14
102:
1.274 brouard 103: Revision 1.273 2017/06/27 11:06:02 brouard
104: Summary: More documentation on projections
105:
1.273 brouard 106: Revision 1.272 2017/06/27 10:22:40 brouard
107: Summary: Color of backprojection changed from 6 to 5(yellow)
108:
1.272 brouard 109: Revision 1.271 2017/06/27 10:17:50 brouard
110: Summary: Some bug with rint
111:
1.271 brouard 112: Revision 1.270 2017/05/24 05:45:29 brouard
113: *** empty log message ***
114:
1.270 brouard 115: Revision 1.269 2017/05/23 08:39:25 brouard
116: Summary: Code into subroutine, cleanings
117:
1.269 brouard 118: Revision 1.268 2017/05/18 20:09:32 brouard
119: Summary: backprojection and confidence intervals of backprevalence
120:
1.268 brouard 121: Revision 1.267 2017/05/13 10:25:05 brouard
122: Summary: temporary save for backprojection
123:
1.267 brouard 124: Revision 1.266 2017/05/13 07:26:12 brouard
125: Summary: Version 0.99r13 (improvements and bugs fixed)
126:
1.266 brouard 127: Revision 1.265 2017/04/26 16:22:11 brouard
128: Summary: imach 0.99r13 Some bugs fixed
129:
1.265 brouard 130: Revision 1.264 2017/04/26 06:01:29 brouard
131: Summary: Labels in graphs
132:
1.264 brouard 133: Revision 1.263 2017/04/24 15:23:15 brouard
134: Summary: to save
135:
1.263 brouard 136: Revision 1.262 2017/04/18 16:48:12 brouard
137: *** empty log message ***
138:
1.262 brouard 139: Revision 1.261 2017/04/05 10:14:09 brouard
140: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
141:
1.261 brouard 142: Revision 1.260 2017/04/04 17:46:59 brouard
143: Summary: Gnuplot indexations fixed (humm)
144:
1.260 brouard 145: Revision 1.259 2017/04/04 13:01:16 brouard
146: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
147:
1.259 brouard 148: Revision 1.258 2017/04/03 10:17:47 brouard
149: Summary: Version 0.99r12
150:
151: Some cleanings, conformed with updated documentation.
152:
1.258 brouard 153: Revision 1.257 2017/03/29 16:53:30 brouard
154: Summary: Temp
155:
1.257 brouard 156: Revision 1.256 2017/03/27 05:50:23 brouard
157: Summary: Temporary
158:
1.256 brouard 159: Revision 1.255 2017/03/08 16:02:28 brouard
160: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
161:
1.255 brouard 162: Revision 1.254 2017/03/08 07:13:00 brouard
163: Summary: Fixing data parameter line
164:
1.254 brouard 165: Revision 1.253 2016/12/15 11:59:41 brouard
166: Summary: 0.99 in progress
167:
1.253 brouard 168: Revision 1.252 2016/09/15 21:15:37 brouard
169: *** empty log message ***
170:
1.252 brouard 171: Revision 1.251 2016/09/15 15:01:13 brouard
172: Summary: not working
173:
1.251 brouard 174: Revision 1.250 2016/09/08 16:07:27 brouard
175: Summary: continue
176:
1.250 brouard 177: Revision 1.249 2016/09/07 17:14:18 brouard
178: Summary: Starting values from frequencies
179:
1.249 brouard 180: Revision 1.248 2016/09/07 14:10:18 brouard
181: *** empty log message ***
182:
1.248 brouard 183: Revision 1.247 2016/09/02 11:11:21 brouard
184: *** empty log message ***
185:
1.247 brouard 186: Revision 1.246 2016/09/02 08:49:22 brouard
187: *** empty log message ***
188:
1.246 brouard 189: Revision 1.245 2016/09/02 07:25:01 brouard
190: *** empty log message ***
191:
1.245 brouard 192: Revision 1.244 2016/09/02 07:17:34 brouard
193: *** empty log message ***
194:
1.244 brouard 195: Revision 1.243 2016/09/02 06:45:35 brouard
196: *** empty log message ***
197:
1.243 brouard 198: Revision 1.242 2016/08/30 15:01:20 brouard
199: Summary: Fixing a lots
200:
1.242 brouard 201: Revision 1.241 2016/08/29 17:17:25 brouard
202: Summary: gnuplot problem in Back projection to fix
203:
1.241 brouard 204: Revision 1.240 2016/08/29 07:53:18 brouard
205: Summary: Better
206:
1.240 brouard 207: Revision 1.239 2016/08/26 15:51:03 brouard
208: Summary: Improvement in Powell output in order to copy and paste
209:
210: Author:
211:
1.239 brouard 212: Revision 1.238 2016/08/26 14:23:35 brouard
213: Summary: Starting tests of 0.99
214:
1.238 brouard 215: Revision 1.237 2016/08/26 09:20:19 brouard
216: Summary: to valgrind
217:
1.237 brouard 218: Revision 1.236 2016/08/25 10:50:18 brouard
219: *** empty log message ***
220:
1.236 brouard 221: Revision 1.235 2016/08/25 06:59:23 brouard
222: *** empty log message ***
223:
1.235 brouard 224: Revision 1.234 2016/08/23 16:51:20 brouard
225: *** empty log message ***
226:
1.234 brouard 227: Revision 1.233 2016/08/23 07:40:50 brouard
228: Summary: not working
229:
1.233 brouard 230: Revision 1.232 2016/08/22 14:20:21 brouard
231: Summary: not working
232:
1.232 brouard 233: Revision 1.231 2016/08/22 07:17:15 brouard
234: Summary: not working
235:
1.231 brouard 236: Revision 1.230 2016/08/22 06:55:53 brouard
237: Summary: Not working
238:
1.230 brouard 239: Revision 1.229 2016/07/23 09:45:53 brouard
240: Summary: Completing for func too
241:
1.229 brouard 242: Revision 1.228 2016/07/22 17:45:30 brouard
243: Summary: Fixing some arrays, still debugging
244:
1.227 brouard 245: Revision 1.226 2016/07/12 18:42:34 brouard
246: Summary: temp
247:
1.226 brouard 248: Revision 1.225 2016/07/12 08:40:03 brouard
249: Summary: saving but not running
250:
1.225 brouard 251: Revision 1.224 2016/07/01 13:16:01 brouard
252: Summary: Fixes
253:
1.224 brouard 254: Revision 1.223 2016/02/19 09:23:35 brouard
255: Summary: temporary
256:
1.223 brouard 257: Revision 1.222 2016/02/17 08:14:50 brouard
258: Summary: Probably last 0.98 stable version 0.98r6
259:
1.222 brouard 260: Revision 1.221 2016/02/15 23:35:36 brouard
261: Summary: minor bug
262:
1.220 brouard 263: Revision 1.219 2016/02/15 00:48:12 brouard
264: *** empty log message ***
265:
1.219 brouard 266: Revision 1.218 2016/02/12 11:29:23 brouard
267: Summary: 0.99 Back projections
268:
1.218 brouard 269: Revision 1.217 2015/12/23 17:18:31 brouard
270: Summary: Experimental backcast
271:
1.217 brouard 272: Revision 1.216 2015/12/18 17:32:11 brouard
273: Summary: 0.98r4 Warning and status=-2
274:
275: Version 0.98r4 is now:
276: - displaying an error when status is -1, date of interview unknown and date of death known;
277: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
278: Older changes concerning s=-2, dating from 2005 have been supersed.
279:
1.216 brouard 280: Revision 1.215 2015/12/16 08:52:24 brouard
281: Summary: 0.98r4 working
282:
1.215 brouard 283: Revision 1.214 2015/12/16 06:57:54 brouard
284: Summary: temporary not working
285:
1.214 brouard 286: Revision 1.213 2015/12/11 18:22:17 brouard
287: Summary: 0.98r4
288:
1.213 brouard 289: Revision 1.212 2015/11/21 12:47:24 brouard
290: Summary: minor typo
291:
1.212 brouard 292: Revision 1.211 2015/11/21 12:41:11 brouard
293: Summary: 0.98r3 with some graph of projected cross-sectional
294:
295: Author: Nicolas Brouard
296:
1.211 brouard 297: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 298: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 299: Summary: Adding ftolpl parameter
300: Author: N Brouard
301:
302: We had difficulties to get smoothed confidence intervals. It was due
303: to the period prevalence which wasn't computed accurately. The inner
304: parameter ftolpl is now an outer parameter of the .imach parameter
305: file after estepm. If ftolpl is small 1.e-4 and estepm too,
306: computation are long.
307:
1.209 brouard 308: Revision 1.208 2015/11/17 14:31:57 brouard
309: Summary: temporary
310:
1.208 brouard 311: Revision 1.207 2015/10/27 17:36:57 brouard
312: *** empty log message ***
313:
1.207 brouard 314: Revision 1.206 2015/10/24 07:14:11 brouard
315: *** empty log message ***
316:
1.206 brouard 317: Revision 1.205 2015/10/23 15:50:53 brouard
318: Summary: 0.98r3 some clarification for graphs on likelihood contributions
319:
1.205 brouard 320: Revision 1.204 2015/10/01 16:20:26 brouard
321: Summary: Some new graphs of contribution to likelihood
322:
1.204 brouard 323: Revision 1.203 2015/09/30 17:45:14 brouard
324: Summary: looking at better estimation of the hessian
325:
326: Also a better criteria for convergence to the period prevalence And
327: therefore adding the number of years needed to converge. (The
328: prevalence in any alive state shold sum to one
329:
1.203 brouard 330: Revision 1.202 2015/09/22 19:45:16 brouard
331: Summary: Adding some overall graph on contribution to likelihood. Might change
332:
1.202 brouard 333: Revision 1.201 2015/09/15 17:34:58 brouard
334: Summary: 0.98r0
335:
336: - Some new graphs like suvival functions
337: - Some bugs fixed like model=1+age+V2.
338:
1.201 brouard 339: Revision 1.200 2015/09/09 16:53:55 brouard
340: Summary: Big bug thanks to Flavia
341:
342: Even model=1+age+V2. did not work anymore
343:
1.200 brouard 344: Revision 1.199 2015/09/07 14:09:23 brouard
345: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
346:
1.199 brouard 347: Revision 1.198 2015/09/03 07:14:39 brouard
348: Summary: 0.98q5 Flavia
349:
1.198 brouard 350: Revision 1.197 2015/09/01 18:24:39 brouard
351: *** empty log message ***
352:
1.197 brouard 353: Revision 1.196 2015/08/18 23:17:52 brouard
354: Summary: 0.98q5
355:
1.196 brouard 356: Revision 1.195 2015/08/18 16:28:39 brouard
357: Summary: Adding a hack for testing purpose
358:
359: After reading the title, ftol and model lines, if the comment line has
360: a q, starting with #q, the answer at the end of the run is quit. It
361: permits to run test files in batch with ctest. The former workaround was
362: $ echo q | imach foo.imach
363:
1.195 brouard 364: Revision 1.194 2015/08/18 13:32:00 brouard
365: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
366:
1.194 brouard 367: Revision 1.193 2015/08/04 07:17:42 brouard
368: Summary: 0.98q4
369:
1.193 brouard 370: Revision 1.192 2015/07/16 16:49:02 brouard
371: Summary: Fixing some outputs
372:
1.192 brouard 373: Revision 1.191 2015/07/14 10:00:33 brouard
374: Summary: Some fixes
375:
1.191 brouard 376: Revision 1.190 2015/05/05 08:51:13 brouard
377: Summary: Adding digits in output parameters (7 digits instead of 6)
378:
379: Fix 1+age+.
380:
1.190 brouard 381: Revision 1.189 2015/04/30 14:45:16 brouard
382: Summary: 0.98q2
383:
1.189 brouard 384: Revision 1.188 2015/04/30 08:27:53 brouard
385: *** empty log message ***
386:
1.188 brouard 387: Revision 1.187 2015/04/29 09:11:15 brouard
388: *** empty log message ***
389:
1.187 brouard 390: Revision 1.186 2015/04/23 12:01:52 brouard
391: Summary: V1*age is working now, version 0.98q1
392:
393: Some codes had been disabled in order to simplify and Vn*age was
394: working in the optimization phase, ie, giving correct MLE parameters,
395: but, as usual, outputs were not correct and program core dumped.
396:
1.186 brouard 397: Revision 1.185 2015/03/11 13:26:42 brouard
398: Summary: Inclusion of compile and links command line for Intel Compiler
399:
1.185 brouard 400: Revision 1.184 2015/03/11 11:52:39 brouard
401: Summary: Back from Windows 8. Intel Compiler
402:
1.184 brouard 403: Revision 1.183 2015/03/10 20:34:32 brouard
404: Summary: 0.98q0, trying with directest, mnbrak fixed
405:
406: We use directest instead of original Powell test; probably no
407: incidence on the results, but better justifications;
408: We fixed Numerical Recipes mnbrak routine which was wrong and gave
409: wrong results.
410:
1.183 brouard 411: Revision 1.182 2015/02/12 08:19:57 brouard
412: Summary: Trying to keep directest which seems simpler and more general
413: Author: Nicolas Brouard
414:
1.182 brouard 415: Revision 1.181 2015/02/11 23:22:24 brouard
416: Summary: Comments on Powell added
417:
418: Author:
419:
1.181 brouard 420: Revision 1.180 2015/02/11 17:33:45 brouard
421: Summary: Finishing move from main to function (hpijx and prevalence_limit)
422:
1.180 brouard 423: Revision 1.179 2015/01/04 09:57:06 brouard
424: Summary: back to OS/X
425:
1.179 brouard 426: Revision 1.178 2015/01/04 09:35:48 brouard
427: *** empty log message ***
428:
1.178 brouard 429: Revision 1.177 2015/01/03 18:40:56 brouard
430: Summary: Still testing ilc32 on OSX
431:
1.177 brouard 432: Revision 1.176 2015/01/03 16:45:04 brouard
433: *** empty log message ***
434:
1.176 brouard 435: Revision 1.175 2015/01/03 16:33:42 brouard
436: *** empty log message ***
437:
1.175 brouard 438: Revision 1.174 2015/01/03 16:15:49 brouard
439: Summary: Still in cross-compilation
440:
1.174 brouard 441: Revision 1.173 2015/01/03 12:06:26 brouard
442: Summary: trying to detect cross-compilation
443:
1.173 brouard 444: Revision 1.172 2014/12/27 12:07:47 brouard
445: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
446:
1.172 brouard 447: Revision 1.171 2014/12/23 13:26:59 brouard
448: Summary: Back from Visual C
449:
450: Still problem with utsname.h on Windows
451:
1.171 brouard 452: Revision 1.170 2014/12/23 11:17:12 brouard
453: Summary: Cleaning some \%% back to %%
454:
455: The escape was mandatory for a specific compiler (which one?), but too many warnings.
456:
1.170 brouard 457: Revision 1.169 2014/12/22 23:08:31 brouard
458: Summary: 0.98p
459:
460: Outputs some informations on compiler used, OS etc. Testing on different platforms.
461:
1.169 brouard 462: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 463: Summary: update
1.169 brouard 464:
1.168 brouard 465: Revision 1.167 2014/12/22 13:50:56 brouard
466: Summary: Testing uname and compiler version and if compiled 32 or 64
467:
468: Testing on Linux 64
469:
1.167 brouard 470: Revision 1.166 2014/12/22 11:40:47 brouard
471: *** empty log message ***
472:
1.166 brouard 473: Revision 1.165 2014/12/16 11:20:36 brouard
474: Summary: After compiling on Visual C
475:
476: * imach.c (Module): Merging 1.61 to 1.162
477:
1.165 brouard 478: Revision 1.164 2014/12/16 10:52:11 brouard
479: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
480:
481: * imach.c (Module): Merging 1.61 to 1.162
482:
1.164 brouard 483: Revision 1.163 2014/12/16 10:30:11 brouard
484: * imach.c (Module): Merging 1.61 to 1.162
485:
1.163 brouard 486: Revision 1.162 2014/09/25 11:43:39 brouard
487: Summary: temporary backup 0.99!
488:
1.162 brouard 489: Revision 1.1 2014/09/16 11:06:58 brouard
490: Summary: With some code (wrong) for nlopt
491:
492: Author:
493:
494: Revision 1.161 2014/09/15 20:41:41 brouard
495: Summary: Problem with macro SQR on Intel compiler
496:
1.161 brouard 497: Revision 1.160 2014/09/02 09:24:05 brouard
498: *** empty log message ***
499:
1.160 brouard 500: Revision 1.159 2014/09/01 10:34:10 brouard
501: Summary: WIN32
502: Author: Brouard
503:
1.159 brouard 504: Revision 1.158 2014/08/27 17:11:51 brouard
505: *** empty log message ***
506:
1.158 brouard 507: Revision 1.157 2014/08/27 16:26:55 brouard
508: Summary: Preparing windows Visual studio version
509: Author: Brouard
510:
511: In order to compile on Visual studio, time.h is now correct and time_t
512: and tm struct should be used. difftime should be used but sometimes I
513: just make the differences in raw time format (time(&now).
514: Trying to suppress #ifdef LINUX
515: Add xdg-open for __linux in order to open default browser.
516:
1.157 brouard 517: Revision 1.156 2014/08/25 20:10:10 brouard
518: *** empty log message ***
519:
1.156 brouard 520: Revision 1.155 2014/08/25 18:32:34 brouard
521: Summary: New compile, minor changes
522: Author: Brouard
523:
1.155 brouard 524: Revision 1.154 2014/06/20 17:32:08 brouard
525: Summary: Outputs now all graphs of convergence to period prevalence
526:
1.154 brouard 527: Revision 1.153 2014/06/20 16:45:46 brouard
528: Summary: If 3 live state, convergence to period prevalence on same graph
529: Author: Brouard
530:
1.153 brouard 531: Revision 1.152 2014/06/18 17:54:09 brouard
532: Summary: open browser, use gnuplot on same dir than imach if not found in the path
533:
1.152 brouard 534: Revision 1.151 2014/06/18 16:43:30 brouard
535: *** empty log message ***
536:
1.151 brouard 537: Revision 1.150 2014/06/18 16:42:35 brouard
538: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
539: Author: brouard
540:
1.150 brouard 541: Revision 1.149 2014/06/18 15:51:14 brouard
542: Summary: Some fixes in parameter files errors
543: Author: Nicolas Brouard
544:
1.149 brouard 545: Revision 1.148 2014/06/17 17:38:48 brouard
546: Summary: Nothing new
547: Author: Brouard
548:
549: Just a new packaging for OS/X version 0.98nS
550:
1.148 brouard 551: Revision 1.147 2014/06/16 10:33:11 brouard
552: *** empty log message ***
553:
1.147 brouard 554: Revision 1.146 2014/06/16 10:20:28 brouard
555: Summary: Merge
556: Author: Brouard
557:
558: Merge, before building revised version.
559:
1.146 brouard 560: Revision 1.145 2014/06/10 21:23:15 brouard
561: Summary: Debugging with valgrind
562: Author: Nicolas Brouard
563:
564: Lot of changes in order to output the results with some covariates
565: After the Edimburgh REVES conference 2014, it seems mandatory to
566: improve the code.
567: No more memory valgrind error but a lot has to be done in order to
568: continue the work of splitting the code into subroutines.
569: Also, decodemodel has been improved. Tricode is still not
570: optimal. nbcode should be improved. Documentation has been added in
571: the source code.
572:
1.144 brouard 573: Revision 1.143 2014/01/26 09:45:38 brouard
574: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
575:
576: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
577: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
578:
1.143 brouard 579: Revision 1.142 2014/01/26 03:57:36 brouard
580: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
581:
582: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
583:
1.142 brouard 584: Revision 1.141 2014/01/26 02:42:01 brouard
585: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
586:
1.141 brouard 587: Revision 1.140 2011/09/02 10:37:54 brouard
588: Summary: times.h is ok with mingw32 now.
589:
1.140 brouard 590: Revision 1.139 2010/06/14 07:50:17 brouard
591: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
592: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
593:
1.139 brouard 594: Revision 1.138 2010/04/30 18:19:40 brouard
595: *** empty log message ***
596:
1.138 brouard 597: Revision 1.137 2010/04/29 18:11:38 brouard
598: (Module): Checking covariates for more complex models
599: than V1+V2. A lot of change to be done. Unstable.
600:
1.137 brouard 601: Revision 1.136 2010/04/26 20:30:53 brouard
602: (Module): merging some libgsl code. Fixing computation
603: of likelione (using inter/intrapolation if mle = 0) in order to
604: get same likelihood as if mle=1.
605: Some cleaning of code and comments added.
606:
1.136 brouard 607: Revision 1.135 2009/10/29 15:33:14 brouard
608: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
609:
1.135 brouard 610: Revision 1.134 2009/10/29 13:18:53 brouard
611: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
612:
1.134 brouard 613: Revision 1.133 2009/07/06 10:21:25 brouard
614: just nforces
615:
1.133 brouard 616: Revision 1.132 2009/07/06 08:22:05 brouard
617: Many tings
618:
1.132 brouard 619: Revision 1.131 2009/06/20 16:22:47 brouard
620: Some dimensions resccaled
621:
1.131 brouard 622: Revision 1.130 2009/05/26 06:44:34 brouard
623: (Module): Max Covariate is now set to 20 instead of 8. A
624: lot of cleaning with variables initialized to 0. Trying to make
625: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
626:
1.130 brouard 627: Revision 1.129 2007/08/31 13:49:27 lievre
628: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
629:
1.129 lievre 630: Revision 1.128 2006/06/30 13:02:05 brouard
631: (Module): Clarifications on computing e.j
632:
1.128 brouard 633: Revision 1.127 2006/04/28 18:11:50 brouard
634: (Module): Yes the sum of survivors was wrong since
635: imach-114 because nhstepm was no more computed in the age
636: loop. Now we define nhstepma in the age loop.
637: (Module): In order to speed up (in case of numerous covariates) we
638: compute health expectancies (without variances) in a first step
639: and then all the health expectancies with variances or standard
640: deviation (needs data from the Hessian matrices) which slows the
641: computation.
642: In the future we should be able to stop the program is only health
643: expectancies and graph are needed without standard deviations.
644:
1.127 brouard 645: Revision 1.126 2006/04/28 17:23:28 brouard
646: (Module): Yes the sum of survivors was wrong since
647: imach-114 because nhstepm was no more computed in the age
648: loop. Now we define nhstepma in the age loop.
649: Version 0.98h
650:
1.126 brouard 651: Revision 1.125 2006/04/04 15:20:31 lievre
652: Errors in calculation of health expectancies. Age was not initialized.
653: Forecasting file added.
654:
655: Revision 1.124 2006/03/22 17:13:53 lievre
656: Parameters are printed with %lf instead of %f (more numbers after the comma).
657: The log-likelihood is printed in the log file
658:
659: Revision 1.123 2006/03/20 10:52:43 brouard
660: * imach.c (Module): <title> changed, corresponds to .htm file
661: name. <head> headers where missing.
662:
663: * imach.c (Module): Weights can have a decimal point as for
664: English (a comma might work with a correct LC_NUMERIC environment,
665: otherwise the weight is truncated).
666: Modification of warning when the covariates values are not 0 or
667: 1.
668: Version 0.98g
669:
670: Revision 1.122 2006/03/20 09:45:41 brouard
671: (Module): Weights can have a decimal point as for
672: English (a comma might work with a correct LC_NUMERIC environment,
673: otherwise the weight is truncated).
674: Modification of warning when the covariates values are not 0 or
675: 1.
676: Version 0.98g
677:
678: Revision 1.121 2006/03/16 17:45:01 lievre
679: * imach.c (Module): Comments concerning covariates added
680:
681: * imach.c (Module): refinements in the computation of lli if
682: status=-2 in order to have more reliable computation if stepm is
683: not 1 month. Version 0.98f
684:
685: Revision 1.120 2006/03/16 15:10:38 lievre
686: (Module): refinements in the computation of lli if
687: status=-2 in order to have more reliable computation if stepm is
688: not 1 month. Version 0.98f
689:
690: Revision 1.119 2006/03/15 17:42:26 brouard
691: (Module): Bug if status = -2, the loglikelihood was
692: computed as likelihood omitting the logarithm. Version O.98e
693:
694: Revision 1.118 2006/03/14 18:20:07 brouard
695: (Module): varevsij Comments added explaining the second
696: table of variances if popbased=1 .
697: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
698: (Module): Function pstamp added
699: (Module): Version 0.98d
700:
701: Revision 1.117 2006/03/14 17:16:22 brouard
702: (Module): varevsij Comments added explaining the second
703: table of variances if popbased=1 .
704: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
705: (Module): Function pstamp added
706: (Module): Version 0.98d
707:
708: Revision 1.116 2006/03/06 10:29:27 brouard
709: (Module): Variance-covariance wrong links and
710: varian-covariance of ej. is needed (Saito).
711:
712: Revision 1.115 2006/02/27 12:17:45 brouard
713: (Module): One freematrix added in mlikeli! 0.98c
714:
715: Revision 1.114 2006/02/26 12:57:58 brouard
716: (Module): Some improvements in processing parameter
717: filename with strsep.
718:
719: Revision 1.113 2006/02/24 14:20:24 brouard
720: (Module): Memory leaks checks with valgrind and:
721: datafile was not closed, some imatrix were not freed and on matrix
722: allocation too.
723:
724: Revision 1.112 2006/01/30 09:55:26 brouard
725: (Module): Back to gnuplot.exe instead of wgnuplot.exe
726:
727: Revision 1.111 2006/01/25 20:38:18 brouard
728: (Module): Lots of cleaning and bugs added (Gompertz)
729: (Module): Comments can be added in data file. Missing date values
730: can be a simple dot '.'.
731:
732: Revision 1.110 2006/01/25 00:51:50 brouard
733: (Module): Lots of cleaning and bugs added (Gompertz)
734:
735: Revision 1.109 2006/01/24 19:37:15 brouard
736: (Module): Comments (lines starting with a #) are allowed in data.
737:
738: Revision 1.108 2006/01/19 18:05:42 lievre
739: Gnuplot problem appeared...
740: To be fixed
741:
742: Revision 1.107 2006/01/19 16:20:37 brouard
743: Test existence of gnuplot in imach path
744:
745: Revision 1.106 2006/01/19 13:24:36 brouard
746: Some cleaning and links added in html output
747:
748: Revision 1.105 2006/01/05 20:23:19 lievre
749: *** empty log message ***
750:
751: Revision 1.104 2005/09/30 16:11:43 lievre
752: (Module): sump fixed, loop imx fixed, and simplifications.
753: (Module): If the status is missing at the last wave but we know
754: that the person is alive, then we can code his/her status as -2
755: (instead of missing=-1 in earlier versions) and his/her
756: contributions to the likelihood is 1 - Prob of dying from last
757: health status (= 1-p13= p11+p12 in the easiest case of somebody in
758: the healthy state at last known wave). Version is 0.98
759:
760: Revision 1.103 2005/09/30 15:54:49 lievre
761: (Module): sump fixed, loop imx fixed, and simplifications.
762:
763: Revision 1.102 2004/09/15 17:31:30 brouard
764: Add the possibility to read data file including tab characters.
765:
766: Revision 1.101 2004/09/15 10:38:38 brouard
767: Fix on curr_time
768:
769: Revision 1.100 2004/07/12 18:29:06 brouard
770: Add version for Mac OS X. Just define UNIX in Makefile
771:
772: Revision 1.99 2004/06/05 08:57:40 brouard
773: *** empty log message ***
774:
775: Revision 1.98 2004/05/16 15:05:56 brouard
776: New version 0.97 . First attempt to estimate force of mortality
777: directly from the data i.e. without the need of knowing the health
778: state at each age, but using a Gompertz model: log u =a + b*age .
779: This is the basic analysis of mortality and should be done before any
780: other analysis, in order to test if the mortality estimated from the
781: cross-longitudinal survey is different from the mortality estimated
782: from other sources like vital statistic data.
783:
784: The same imach parameter file can be used but the option for mle should be -3.
785:
1.133 brouard 786: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 787: former routines in order to include the new code within the former code.
788:
789: The output is very simple: only an estimate of the intercept and of
790: the slope with 95% confident intervals.
791:
792: Current limitations:
793: A) Even if you enter covariates, i.e. with the
794: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
795: B) There is no computation of Life Expectancy nor Life Table.
796:
797: Revision 1.97 2004/02/20 13:25:42 lievre
798: Version 0.96d. Population forecasting command line is (temporarily)
799: suppressed.
800:
801: Revision 1.96 2003/07/15 15:38:55 brouard
802: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
803: rewritten within the same printf. Workaround: many printfs.
804:
805: Revision 1.95 2003/07/08 07:54:34 brouard
806: * imach.c (Repository):
807: (Repository): Using imachwizard code to output a more meaningful covariance
808: matrix (cov(a12,c31) instead of numbers.
809:
810: Revision 1.94 2003/06/27 13:00:02 brouard
811: Just cleaning
812:
813: Revision 1.93 2003/06/25 16:33:55 brouard
814: (Module): On windows (cygwin) function asctime_r doesn't
815: exist so I changed back to asctime which exists.
816: (Module): Version 0.96b
817:
818: Revision 1.92 2003/06/25 16:30:45 brouard
819: (Module): On windows (cygwin) function asctime_r doesn't
820: exist so I changed back to asctime which exists.
821:
822: Revision 1.91 2003/06/25 15:30:29 brouard
823: * imach.c (Repository): Duplicated warning errors corrected.
824: (Repository): Elapsed time after each iteration is now output. It
825: helps to forecast when convergence will be reached. Elapsed time
826: is stamped in powell. We created a new html file for the graphs
827: concerning matrix of covariance. It has extension -cov.htm.
828:
829: Revision 1.90 2003/06/24 12:34:15 brouard
830: (Module): Some bugs corrected for windows. Also, when
831: mle=-1 a template is output in file "or"mypar.txt with the design
832: of the covariance matrix to be input.
833:
834: Revision 1.89 2003/06/24 12:30:52 brouard
835: (Module): Some bugs corrected for windows. Also, when
836: mle=-1 a template is output in file "or"mypar.txt with the design
837: of the covariance matrix to be input.
838:
839: Revision 1.88 2003/06/23 17:54:56 brouard
840: * 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.
841:
842: Revision 1.87 2003/06/18 12:26:01 brouard
843: Version 0.96
844:
845: Revision 1.86 2003/06/17 20:04:08 brouard
846: (Module): Change position of html and gnuplot routines and added
847: routine fileappend.
848:
849: Revision 1.85 2003/06/17 13:12:43 brouard
850: * imach.c (Repository): Check when date of death was earlier that
851: current date of interview. It may happen when the death was just
852: prior to the death. In this case, dh was negative and likelihood
853: was wrong (infinity). We still send an "Error" but patch by
854: assuming that the date of death was just one stepm after the
855: interview.
856: (Repository): Because some people have very long ID (first column)
857: we changed int to long in num[] and we added a new lvector for
858: memory allocation. But we also truncated to 8 characters (left
859: truncation)
860: (Repository): No more line truncation errors.
861:
862: Revision 1.84 2003/06/13 21:44:43 brouard
863: * imach.c (Repository): Replace "freqsummary" at a correct
864: place. It differs from routine "prevalence" which may be called
865: many times. Probs is memory consuming and must be used with
866: parcimony.
867: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
868:
869: Revision 1.83 2003/06/10 13:39:11 lievre
870: *** empty log message ***
871:
872: Revision 1.82 2003/06/05 15:57:20 brouard
873: Add log in imach.c and fullversion number is now printed.
874:
875: */
876: /*
877: Interpolated Markov Chain
878:
879: Short summary of the programme:
880:
1.227 brouard 881: This program computes Healthy Life Expectancies or State-specific
882: (if states aren't health statuses) Expectancies from
883: cross-longitudinal data. Cross-longitudinal data consist in:
884:
885: -1- a first survey ("cross") where individuals from different ages
886: are interviewed on their health status or degree of disability (in
887: the case of a health survey which is our main interest)
888:
889: -2- at least a second wave of interviews ("longitudinal") which
890: measure each change (if any) in individual health status. Health
891: expectancies are computed from the time spent in each health state
892: according to a model. More health states you consider, more time is
893: necessary to reach the Maximum Likelihood of the parameters involved
894: in the model. The simplest model is the multinomial logistic model
895: where pij is the probability to be observed in state j at the second
896: wave conditional to be observed in state i at the first
897: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
898: etc , where 'age' is age and 'sex' is a covariate. If you want to
899: have a more complex model than "constant and age", you should modify
900: the program where the markup *Covariates have to be included here
901: again* invites you to do it. More covariates you add, slower the
1.126 brouard 902: convergence.
903:
904: The advantage of this computer programme, compared to a simple
905: multinomial logistic model, is clear when the delay between waves is not
906: identical for each individual. Also, if a individual missed an
907: intermediate interview, the information is lost, but taken into
908: account using an interpolation or extrapolation.
909:
910: hPijx is the probability to be observed in state i at age x+h
911: conditional to the observed state i at age x. The delay 'h' can be
912: split into an exact number (nh*stepm) of unobserved intermediate
913: states. This elementary transition (by month, quarter,
914: semester or year) is modelled as a multinomial logistic. The hPx
915: matrix is simply the matrix product of nh*stepm elementary matrices
916: and the contribution of each individual to the likelihood is simply
917: hPijx.
918:
919: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 920: of the life expectancies. It also computes the period (stable) prevalence.
921:
922: Back prevalence and projections:
1.227 brouard 923:
924: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
925: double agemaxpar, double ftolpl, int *ncvyearp, double
926: dateprev1,double dateprev2, int firstpass, int lastpass, int
927: mobilavproj)
928:
929: Computes the back prevalence limit for any combination of
930: covariate values k at any age between ageminpar and agemaxpar and
931: returns it in **bprlim. In the loops,
932:
933: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
934: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
935:
936: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 937: Computes for any combination of covariates k and any age between bage and fage
938: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
939: oldm=oldms;savm=savms;
1.227 brouard 940:
1.267 brouard 941: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 942: Computes the transition matrix starting at age 'age' over
943: 'nhstepm*hstepm*stepm' months (i.e. until
944: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 945: nhstepm*hstepm matrices.
946:
947: Returns p3mat[i][j][h] after calling
948: p3mat[i][j][h]=matprod2(newm,
949: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
950: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
951: oldm);
1.226 brouard 952:
953: Important routines
954:
955: - func (or funcone), computes logit (pij) distinguishing
956: o fixed variables (single or product dummies or quantitative);
957: o varying variables by:
958: (1) wave (single, product dummies, quantitative),
959: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
960: % fixed dummy (treated) or quantitative (not done because time-consuming);
961: % varying dummy (not done) or quantitative (not done);
962: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
963: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
964: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
965: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
966: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 967:
1.226 brouard 968:
969:
1.133 brouard 970: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
971: Institut national d'études démographiques, Paris.
1.126 brouard 972: This software have been partly granted by Euro-REVES, a concerted action
973: from the European Union.
974: It is copyrighted identically to a GNU software product, ie programme and
975: software can be distributed freely for non commercial use. Latest version
976: can be accessed at http://euroreves.ined.fr/imach .
977:
978: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
979: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
980:
981: **********************************************************************/
982: /*
983: main
984: read parameterfile
985: read datafile
986: concatwav
987: freqsummary
988: if (mle >= 1)
989: mlikeli
990: print results files
991: if mle==1
992: computes hessian
993: read end of parameter file: agemin, agemax, bage, fage, estepm
994: begin-prev-date,...
995: open gnuplot file
996: open html file
1.145 brouard 997: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
998: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
999: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1000: freexexit2 possible for memory heap.
1001:
1002: h Pij x | pij_nom ficrestpij
1003: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1004: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1005: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1006:
1007: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1008: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1009: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1010: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1011: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1012:
1.126 brouard 1013: forecasting if prevfcast==1 prevforecast call prevalence()
1014: health expectancies
1015: Variance-covariance of DFLE
1016: prevalence()
1017: movingaverage()
1018: varevsij()
1019: if popbased==1 varevsij(,popbased)
1020: total life expectancies
1021: Variance of period (stable) prevalence
1022: end
1023: */
1024:
1.187 brouard 1025: /* #define DEBUG */
1026: /* #define DEBUGBRENT */
1.203 brouard 1027: /* #define DEBUGLINMIN */
1028: /* #define DEBUGHESS */
1029: #define DEBUGHESSIJ
1.224 brouard 1030: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1031: #define POWELL /* Instead of NLOPT */
1.224 brouard 1032: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1033: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1034: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1035:
1036: #include <math.h>
1037: #include <stdio.h>
1038: #include <stdlib.h>
1039: #include <string.h>
1.226 brouard 1040: #include <ctype.h>
1.159 brouard 1041:
1042: #ifdef _WIN32
1043: #include <io.h>
1.172 brouard 1044: #include <windows.h>
1045: #include <tchar.h>
1.159 brouard 1046: #else
1.126 brouard 1047: #include <unistd.h>
1.159 brouard 1048: #endif
1.126 brouard 1049:
1050: #include <limits.h>
1051: #include <sys/types.h>
1.171 brouard 1052:
1053: #if defined(__GNUC__)
1054: #include <sys/utsname.h> /* Doesn't work on Windows */
1055: #endif
1056:
1.126 brouard 1057: #include <sys/stat.h>
1058: #include <errno.h>
1.159 brouard 1059: /* extern int errno; */
1.126 brouard 1060:
1.157 brouard 1061: /* #ifdef LINUX */
1062: /* #include <time.h> */
1063: /* #include "timeval.h" */
1064: /* #else */
1065: /* #include <sys/time.h> */
1066: /* #endif */
1067:
1.126 brouard 1068: #include <time.h>
1069:
1.136 brouard 1070: #ifdef GSL
1071: #include <gsl/gsl_errno.h>
1072: #include <gsl/gsl_multimin.h>
1073: #endif
1074:
1.167 brouard 1075:
1.162 brouard 1076: #ifdef NLOPT
1077: #include <nlopt.h>
1078: typedef struct {
1079: double (* function)(double [] );
1080: } myfunc_data ;
1081: #endif
1082:
1.126 brouard 1083: /* #include <libintl.h> */
1084: /* #define _(String) gettext (String) */
1085:
1.251 brouard 1086: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1087:
1088: #define GNUPLOTPROGRAM "gnuplot"
1089: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1090: #define FILENAMELENGTH 132
1091:
1092: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1093: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1094:
1.144 brouard 1095: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1096: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1097:
1098: #define NINTERVMAX 8
1.144 brouard 1099: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1100: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1101: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1102: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1103: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1104: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1105: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1106: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1107: /* #define AGESUP 130 */
1.288 brouard 1108: /* #define AGESUP 150 */
1109: #define AGESUP 200
1.268 brouard 1110: #define AGEINF 0
1.218 brouard 1111: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1112: #define AGEBASE 40
1.194 brouard 1113: #define AGEOVERFLOW 1.e20
1.164 brouard 1114: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1115: #ifdef _WIN32
1116: #define DIRSEPARATOR '\\'
1117: #define CHARSEPARATOR "\\"
1118: #define ODIRSEPARATOR '/'
1119: #else
1.126 brouard 1120: #define DIRSEPARATOR '/'
1121: #define CHARSEPARATOR "/"
1122: #define ODIRSEPARATOR '\\'
1123: #endif
1124:
1.302 ! brouard 1125: /* $Id: imach.c,v 1.301 2019/06/04 13:51:20 brouard Exp $ */
1.126 brouard 1126: /* $State: Exp $ */
1.196 brouard 1127: #include "version.h"
1128: char version[]=__IMACH_VERSION__;
1.300 brouard 1129: char copyright[]="May 2019,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020";
1.302 ! brouard 1130: char fullversion[]="$Revision: 1.301 $ $Date: 2019/06/04 13:51:20 $";
1.126 brouard 1131: char strstart[80];
1132: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1133: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1134: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1135: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1136: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1137: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1138: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1139: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1140: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1141: int cptcovprodnoage=0; /**< Number of covariate products without age */
1142: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1143: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1144: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1145: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1146: int nsd=0; /**< Total number of single dummy variables (output) */
1147: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1148: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1149: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1150: int ntveff=0; /**< ntveff number of effective time varying variables */
1151: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1152: int cptcov=0; /* Working variable */
1.290 brouard 1153: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1154: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 ! brouard 1155: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1156: int nlstate=2; /* Number of live states */
1157: int ndeath=1; /* Number of dead states */
1.130 brouard 1158: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1159: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1160: int popbased=0;
1161:
1162: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1163: int maxwav=0; /* Maxim number of waves */
1164: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1165: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1166: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1167: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1168: int mle=1, weightopt=0;
1.126 brouard 1169: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1170: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1171: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1172: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1173: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1174: int selected(int kvar); /* Is covariate kvar selected for printing results */
1175:
1.130 brouard 1176: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1177: double **matprod2(); /* test */
1.126 brouard 1178: double **oldm, **newm, **savm; /* Working pointers to matrices */
1179: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1180: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1181:
1.136 brouard 1182: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1183: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1184: FILE *ficlog, *ficrespow;
1.130 brouard 1185: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1186: double fretone; /* Only one call to likelihood */
1.130 brouard 1187: long ipmx=0; /* Number of contributions */
1.126 brouard 1188: double sw; /* Sum of weights */
1189: char filerespow[FILENAMELENGTH];
1190: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1191: FILE *ficresilk;
1192: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1193: FILE *ficresprobmorprev;
1194: FILE *fichtm, *fichtmcov; /* Html File */
1195: FILE *ficreseij;
1196: char filerese[FILENAMELENGTH];
1197: FILE *ficresstdeij;
1198: char fileresstde[FILENAMELENGTH];
1199: FILE *ficrescveij;
1200: char filerescve[FILENAMELENGTH];
1201: FILE *ficresvij;
1202: char fileresv[FILENAMELENGTH];
1.269 brouard 1203:
1.126 brouard 1204: char title[MAXLINE];
1.234 brouard 1205: char model[MAXLINE]; /**< The model line */
1.217 brouard 1206: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1207: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1208: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1209: char command[FILENAMELENGTH];
1210: int outcmd=0;
1211:
1.217 brouard 1212: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1213: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1214: char filelog[FILENAMELENGTH]; /* Log file */
1215: char filerest[FILENAMELENGTH];
1216: char fileregp[FILENAMELENGTH];
1217: char popfile[FILENAMELENGTH];
1218:
1219: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1220:
1.157 brouard 1221: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1222: /* struct timezone tzp; */
1223: /* extern int gettimeofday(); */
1224: struct tm tml, *gmtime(), *localtime();
1225:
1226: extern time_t time();
1227:
1228: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1229: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1230: struct tm tm;
1231:
1.126 brouard 1232: char strcurr[80], strfor[80];
1233:
1234: char *endptr;
1235: long lval;
1236: double dval;
1237:
1238: #define NR_END 1
1239: #define FREE_ARG char*
1240: #define FTOL 1.0e-10
1241:
1242: #define NRANSI
1.240 brouard 1243: #define ITMAX 200
1244: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1245:
1246: #define TOL 2.0e-4
1247:
1248: #define CGOLD 0.3819660
1249: #define ZEPS 1.0e-10
1250: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1251:
1252: #define GOLD 1.618034
1253: #define GLIMIT 100.0
1254: #define TINY 1.0e-20
1255:
1256: static double maxarg1,maxarg2;
1257: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1258: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1259:
1260: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1261: #define rint(a) floor(a+0.5)
1.166 brouard 1262: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1263: #define mytinydouble 1.0e-16
1.166 brouard 1264: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1265: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1266: /* static double dsqrarg; */
1267: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1268: static double sqrarg;
1269: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1270: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1271: int agegomp= AGEGOMP;
1272:
1273: int imx;
1274: int stepm=1;
1275: /* Stepm, step in month: minimum step interpolation*/
1276:
1277: int estepm;
1278: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1279:
1280: int m,nb;
1281: long *num;
1.197 brouard 1282: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1283: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1284: covariate for which somebody answered excluding
1285: undefined. Usually 2: 0 and 1. */
1286: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1287: covariate for which somebody answered including
1288: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1289: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1290: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1291: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1292: double *ageexmed,*agecens;
1293: double dateintmean=0;
1.296 brouard 1294: double anprojd, mprojd, jprojd; /* For eventual projections */
1295: double anprojf, mprojf, jprojf;
1.126 brouard 1296:
1.296 brouard 1297: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1298: double anbackf, mbackf, jbackf;
1299: double jintmean,mintmean,aintmean;
1.126 brouard 1300: double *weight;
1301: int **s; /* Status */
1.141 brouard 1302: double *agedc;
1.145 brouard 1303: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1304: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1305: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1306: double **coqvar; /* Fixed quantitative covariate nqv */
1307: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1308: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1309: double idx;
1310: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1311: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1312: /*k 1 2 3 4 5 6 7 8 9 */
1313: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1314: /* Tndvar[k] 1 2 3 4 5 */
1315: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1316: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1317: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1318: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1319: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1320: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1321: /* Tprod[i]=k 4 7 */
1322: /* Tage[i]=k 5 8 */
1323: /* */
1324: /* Type */
1325: /* V 1 2 3 4 5 */
1326: /* F F V V V */
1327: /* D Q D D Q */
1328: /* */
1329: int *TvarsD;
1330: int *TvarsDind;
1331: int *TvarsQ;
1332: int *TvarsQind;
1333:
1.235 brouard 1334: #define MAXRESULTLINES 10
1335: int nresult=0;
1.258 brouard 1336: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1337: int TKresult[MAXRESULTLINES];
1.237 brouard 1338: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1339: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1340: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1341: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1342: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1343: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1344:
1.234 brouard 1345: /* 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 1346: 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 */
1347: 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 */
1348: 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 */
1349: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1350: 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 */
1351: 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 1352: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1353: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1354: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1355: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1356: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1357: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1358: 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 */
1359: 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 */
1360:
1.230 brouard 1361: int *Tvarsel; /**< Selected covariates for output */
1362: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1363: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1364: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1365: 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 1366: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1367: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1368: int *Tage;
1.227 brouard 1369: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1370: 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 1371: 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*/
1372: 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 1373: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1374: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1375: int **Tvard;
1376: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1377: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1378: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1379: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1380: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1381: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1382: double *lsurv, *lpop, *tpop;
1383:
1.231 brouard 1384: #define FD 1; /* Fixed dummy covariate */
1385: #define FQ 2; /* Fixed quantitative covariate */
1386: #define FP 3; /* Fixed product covariate */
1387: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1388: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1389: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1390: #define VD 10; /* Varying dummy covariate */
1391: #define VQ 11; /* Varying quantitative covariate */
1392: #define VP 12; /* Varying product covariate */
1393: #define VPDD 13; /* Varying product dummy*dummy covariate */
1394: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1395: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1396: #define APFD 16; /* Age product * fixed dummy covariate */
1397: #define APFQ 17; /* Age product * fixed quantitative covariate */
1398: #define APVD 18; /* Age product * varying dummy covariate */
1399: #define APVQ 19; /* Age product * varying quantitative covariate */
1400:
1401: #define FTYPE 1; /* Fixed covariate */
1402: #define VTYPE 2; /* Varying covariate (loop in wave) */
1403: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1404:
1405: struct kmodel{
1406: int maintype; /* main type */
1407: int subtype; /* subtype */
1408: };
1409: struct kmodel modell[NCOVMAX];
1410:
1.143 brouard 1411: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1412: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1413:
1414: /**************** split *************************/
1415: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1416: {
1417: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1418: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1419: */
1420: char *ss; /* pointer */
1.186 brouard 1421: int l1=0, l2=0; /* length counters */
1.126 brouard 1422:
1423: l1 = strlen(path ); /* length of path */
1424: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1425: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1426: if ( ss == NULL ) { /* no directory, so determine current directory */
1427: strcpy( name, path ); /* we got the fullname name because no directory */
1428: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1429: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1430: /* get current working directory */
1431: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1432: #ifdef WIN32
1433: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1434: #else
1435: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1436: #endif
1.126 brouard 1437: return( GLOCK_ERROR_GETCWD );
1438: }
1439: /* got dirc from getcwd*/
1440: printf(" DIRC = %s \n",dirc);
1.205 brouard 1441: } else { /* strip directory from path */
1.126 brouard 1442: ss++; /* after this, the filename */
1443: l2 = strlen( ss ); /* length of filename */
1444: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1445: strcpy( name, ss ); /* save file name */
1446: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1447: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1448: printf(" DIRC2 = %s \n",dirc);
1449: }
1450: /* We add a separator at the end of dirc if not exists */
1451: l1 = strlen( dirc ); /* length of directory */
1452: if( dirc[l1-1] != DIRSEPARATOR ){
1453: dirc[l1] = DIRSEPARATOR;
1454: dirc[l1+1] = 0;
1455: printf(" DIRC3 = %s \n",dirc);
1456: }
1457: ss = strrchr( name, '.' ); /* find last / */
1458: if (ss >0){
1459: ss++;
1460: strcpy(ext,ss); /* save extension */
1461: l1= strlen( name);
1462: l2= strlen(ss)+1;
1463: strncpy( finame, name, l1-l2);
1464: finame[l1-l2]= 0;
1465: }
1466:
1467: return( 0 ); /* we're done */
1468: }
1469:
1470:
1471: /******************************************/
1472:
1473: void replace_back_to_slash(char *s, char*t)
1474: {
1475: int i;
1476: int lg=0;
1477: i=0;
1478: lg=strlen(t);
1479: for(i=0; i<= lg; i++) {
1480: (s[i] = t[i]);
1481: if (t[i]== '\\') s[i]='/';
1482: }
1483: }
1484:
1.132 brouard 1485: char *trimbb(char *out, char *in)
1.137 brouard 1486: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1487: char *s;
1488: s=out;
1489: while (*in != '\0'){
1.137 brouard 1490: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1491: in++;
1492: }
1493: *out++ = *in++;
1494: }
1495: *out='\0';
1496: return s;
1497: }
1498:
1.187 brouard 1499: /* char *substrchaine(char *out, char *in, char *chain) */
1500: /* { */
1501: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1502: /* char *s, *t; */
1503: /* t=in;s=out; */
1504: /* while ((*in != *chain) && (*in != '\0')){ */
1505: /* *out++ = *in++; */
1506: /* } */
1507:
1508: /* /\* *in matches *chain *\/ */
1509: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1510: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1511: /* } */
1512: /* in--; chain--; */
1513: /* while ( (*in != '\0')){ */
1514: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1515: /* *out++ = *in++; */
1516: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1517: /* } */
1518: /* *out='\0'; */
1519: /* out=s; */
1520: /* return out; */
1521: /* } */
1522: char *substrchaine(char *out, char *in, char *chain)
1523: {
1524: /* Substract chain 'chain' from 'in', return and output 'out' */
1525: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1526:
1527: char *strloc;
1528:
1529: strcpy (out, in);
1530: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1531: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1532: if(strloc != NULL){
1533: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1534: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1535: /* strcpy (strloc, strloc +strlen(chain));*/
1536: }
1537: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1538: return out;
1539: }
1540:
1541:
1.145 brouard 1542: char *cutl(char *blocc, char *alocc, char *in, char occ)
1543: {
1.187 brouard 1544: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1545: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1546: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1547: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1548: */
1.160 brouard 1549: char *s, *t;
1.145 brouard 1550: t=in;s=in;
1551: while ((*in != occ) && (*in != '\0')){
1552: *alocc++ = *in++;
1553: }
1554: if( *in == occ){
1555: *(alocc)='\0';
1556: s=++in;
1557: }
1558:
1559: if (s == t) {/* occ not found */
1560: *(alocc-(in-s))='\0';
1561: in=s;
1562: }
1563: while ( *in != '\0'){
1564: *blocc++ = *in++;
1565: }
1566:
1567: *blocc='\0';
1568: return t;
1569: }
1.137 brouard 1570: char *cutv(char *blocc, char *alocc, char *in, char occ)
1571: {
1.187 brouard 1572: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1573: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1574: gives blocc="abcdef2ghi" and alocc="j".
1575: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1576: */
1577: char *s, *t;
1578: t=in;s=in;
1579: while (*in != '\0'){
1580: while( *in == occ){
1581: *blocc++ = *in++;
1582: s=in;
1583: }
1584: *blocc++ = *in++;
1585: }
1586: if (s == t) /* occ not found */
1587: *(blocc-(in-s))='\0';
1588: else
1589: *(blocc-(in-s)-1)='\0';
1590: in=s;
1591: while ( *in != '\0'){
1592: *alocc++ = *in++;
1593: }
1594:
1595: *alocc='\0';
1596: return s;
1597: }
1598:
1.126 brouard 1599: int nbocc(char *s, char occ)
1600: {
1601: int i,j=0;
1602: int lg=20;
1603: i=0;
1604: lg=strlen(s);
1605: for(i=0; i<= lg; i++) {
1.234 brouard 1606: if (s[i] == occ ) j++;
1.126 brouard 1607: }
1608: return j;
1609: }
1610:
1.137 brouard 1611: /* void cutv(char *u,char *v, char*t, char occ) */
1612: /* { */
1613: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1614: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1615: /* gives u="abcdef2ghi" and v="j" *\/ */
1616: /* int i,lg,j,p=0; */
1617: /* i=0; */
1618: /* lg=strlen(t); */
1619: /* for(j=0; j<=lg-1; j++) { */
1620: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1621: /* } */
1.126 brouard 1622:
1.137 brouard 1623: /* for(j=0; j<p; j++) { */
1624: /* (u[j] = t[j]); */
1625: /* } */
1626: /* u[p]='\0'; */
1.126 brouard 1627:
1.137 brouard 1628: /* for(j=0; j<= lg; j++) { */
1629: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1630: /* } */
1631: /* } */
1.126 brouard 1632:
1.160 brouard 1633: #ifdef _WIN32
1634: char * strsep(char **pp, const char *delim)
1635: {
1636: char *p, *q;
1637:
1638: if ((p = *pp) == NULL)
1639: return 0;
1640: if ((q = strpbrk (p, delim)) != NULL)
1641: {
1642: *pp = q + 1;
1643: *q = '\0';
1644: }
1645: else
1646: *pp = 0;
1647: return p;
1648: }
1649: #endif
1650:
1.126 brouard 1651: /********************** nrerror ********************/
1652:
1653: void nrerror(char error_text[])
1654: {
1655: fprintf(stderr,"ERREUR ...\n");
1656: fprintf(stderr,"%s\n",error_text);
1657: exit(EXIT_FAILURE);
1658: }
1659: /*********************** vector *******************/
1660: double *vector(int nl, int nh)
1661: {
1662: double *v;
1663: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1664: if (!v) nrerror("allocation failure in vector");
1665: return v-nl+NR_END;
1666: }
1667:
1668: /************************ free vector ******************/
1669: void free_vector(double*v, int nl, int nh)
1670: {
1671: free((FREE_ARG)(v+nl-NR_END));
1672: }
1673:
1674: /************************ivector *******************************/
1675: int *ivector(long nl,long nh)
1676: {
1677: int *v;
1678: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1679: if (!v) nrerror("allocation failure in ivector");
1680: return v-nl+NR_END;
1681: }
1682:
1683: /******************free ivector **************************/
1684: void free_ivector(int *v, long nl, long nh)
1685: {
1686: free((FREE_ARG)(v+nl-NR_END));
1687: }
1688:
1689: /************************lvector *******************************/
1690: long *lvector(long nl,long nh)
1691: {
1692: long *v;
1693: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1694: if (!v) nrerror("allocation failure in ivector");
1695: return v-nl+NR_END;
1696: }
1697:
1698: /******************free lvector **************************/
1699: void free_lvector(long *v, long nl, long nh)
1700: {
1701: free((FREE_ARG)(v+nl-NR_END));
1702: }
1703:
1704: /******************* imatrix *******************************/
1705: int **imatrix(long nrl, long nrh, long ncl, long nch)
1706: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1707: {
1708: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1709: int **m;
1710:
1711: /* allocate pointers to rows */
1712: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1713: if (!m) nrerror("allocation failure 1 in matrix()");
1714: m += NR_END;
1715: m -= nrl;
1716:
1717:
1718: /* allocate rows and set pointers to them */
1719: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1720: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1721: m[nrl] += NR_END;
1722: m[nrl] -= ncl;
1723:
1724: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1725:
1726: /* return pointer to array of pointers to rows */
1727: return m;
1728: }
1729:
1730: /****************** free_imatrix *************************/
1731: void free_imatrix(m,nrl,nrh,ncl,nch)
1732: int **m;
1733: long nch,ncl,nrh,nrl;
1734: /* free an int matrix allocated by imatrix() */
1735: {
1736: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1737: free((FREE_ARG) (m+nrl-NR_END));
1738: }
1739:
1740: /******************* matrix *******************************/
1741: double **matrix(long nrl, long nrh, long ncl, long nch)
1742: {
1743: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1744: double **m;
1745:
1746: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1747: if (!m) nrerror("allocation failure 1 in matrix()");
1748: m += NR_END;
1749: m -= nrl;
1750:
1751: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1752: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1753: m[nrl] += NR_END;
1754: m[nrl] -= ncl;
1755:
1756: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1757: return m;
1.145 brouard 1758: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1759: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1760: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1761: */
1762: }
1763:
1764: /*************************free matrix ************************/
1765: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1766: {
1767: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1768: free((FREE_ARG)(m+nrl-NR_END));
1769: }
1770:
1771: /******************* ma3x *******************************/
1772: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1773: {
1774: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1775: double ***m;
1776:
1777: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1778: if (!m) nrerror("allocation failure 1 in matrix()");
1779: m += NR_END;
1780: m -= nrl;
1781:
1782: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1783: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1784: m[nrl] += NR_END;
1785: m[nrl] -= ncl;
1786:
1787: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1788:
1789: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1790: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1791: m[nrl][ncl] += NR_END;
1792: m[nrl][ncl] -= nll;
1793: for (j=ncl+1; j<=nch; j++)
1794: m[nrl][j]=m[nrl][j-1]+nlay;
1795:
1796: for (i=nrl+1; i<=nrh; i++) {
1797: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1798: for (j=ncl+1; j<=nch; j++)
1799: m[i][j]=m[i][j-1]+nlay;
1800: }
1801: return m;
1802: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1803: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1804: */
1805: }
1806:
1807: /*************************free ma3x ************************/
1808: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1809: {
1810: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1811: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1812: free((FREE_ARG)(m+nrl-NR_END));
1813: }
1814:
1815: /*************** function subdirf ***********/
1816: char *subdirf(char fileres[])
1817: {
1818: /* Caution optionfilefiname is hidden */
1819: strcpy(tmpout,optionfilefiname);
1820: strcat(tmpout,"/"); /* Add to the right */
1821: strcat(tmpout,fileres);
1822: return tmpout;
1823: }
1824:
1825: /*************** function subdirf2 ***********/
1826: char *subdirf2(char fileres[], char *preop)
1827: {
1828:
1829: /* Caution optionfilefiname is hidden */
1830: strcpy(tmpout,optionfilefiname);
1831: strcat(tmpout,"/");
1832: strcat(tmpout,preop);
1833: strcat(tmpout,fileres);
1834: return tmpout;
1835: }
1836:
1837: /*************** function subdirf3 ***********/
1838: char *subdirf3(char fileres[], char *preop, char *preop2)
1839: {
1840:
1841: /* Caution optionfilefiname is hidden */
1842: strcpy(tmpout,optionfilefiname);
1843: strcat(tmpout,"/");
1844: strcat(tmpout,preop);
1845: strcat(tmpout,preop2);
1846: strcat(tmpout,fileres);
1847: return tmpout;
1848: }
1.213 brouard 1849:
1850: /*************** function subdirfext ***********/
1851: char *subdirfext(char fileres[], char *preop, char *postop)
1852: {
1853:
1854: strcpy(tmpout,preop);
1855: strcat(tmpout,fileres);
1856: strcat(tmpout,postop);
1857: return tmpout;
1858: }
1.126 brouard 1859:
1.213 brouard 1860: /*************** function subdirfext3 ***********/
1861: char *subdirfext3(char fileres[], char *preop, char *postop)
1862: {
1863:
1864: /* Caution optionfilefiname is hidden */
1865: strcpy(tmpout,optionfilefiname);
1866: strcat(tmpout,"/");
1867: strcat(tmpout,preop);
1868: strcat(tmpout,fileres);
1869: strcat(tmpout,postop);
1870: return tmpout;
1871: }
1872:
1.162 brouard 1873: char *asc_diff_time(long time_sec, char ascdiff[])
1874: {
1875: long sec_left, days, hours, minutes;
1876: days = (time_sec) / (60*60*24);
1877: sec_left = (time_sec) % (60*60*24);
1878: hours = (sec_left) / (60*60) ;
1879: sec_left = (sec_left) %(60*60);
1880: minutes = (sec_left) /60;
1881: sec_left = (sec_left) % (60);
1882: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1883: return ascdiff;
1884: }
1885:
1.126 brouard 1886: /***************** f1dim *************************/
1887: extern int ncom;
1888: extern double *pcom,*xicom;
1889: extern double (*nrfunc)(double []);
1890:
1891: double f1dim(double x)
1892: {
1893: int j;
1894: double f;
1895: double *xt;
1896:
1897: xt=vector(1,ncom);
1898: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1899: f=(*nrfunc)(xt);
1900: free_vector(xt,1,ncom);
1901: return f;
1902: }
1903:
1904: /*****************brent *************************/
1905: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1906: {
1907: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1908: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1909: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1910: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1911: * returned function value.
1912: */
1.126 brouard 1913: int iter;
1914: double a,b,d,etemp;
1.159 brouard 1915: double fu=0,fv,fw,fx;
1.164 brouard 1916: double ftemp=0.;
1.126 brouard 1917: double p,q,r,tol1,tol2,u,v,w,x,xm;
1918: double e=0.0;
1919:
1920: a=(ax < cx ? ax : cx);
1921: b=(ax > cx ? ax : cx);
1922: x=w=v=bx;
1923: fw=fv=fx=(*f)(x);
1924: for (iter=1;iter<=ITMAX;iter++) {
1925: xm=0.5*(a+b);
1926: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1927: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1928: printf(".");fflush(stdout);
1929: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1930: #ifdef DEBUGBRENT
1.126 brouard 1931: 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);
1932: 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);
1933: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1934: #endif
1935: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1936: *xmin=x;
1937: return fx;
1938: }
1939: ftemp=fu;
1940: if (fabs(e) > tol1) {
1941: r=(x-w)*(fx-fv);
1942: q=(x-v)*(fx-fw);
1943: p=(x-v)*q-(x-w)*r;
1944: q=2.0*(q-r);
1945: if (q > 0.0) p = -p;
1946: q=fabs(q);
1947: etemp=e;
1948: e=d;
1949: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1950: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1951: else {
1.224 brouard 1952: d=p/q;
1953: u=x+d;
1954: if (u-a < tol2 || b-u < tol2)
1955: d=SIGN(tol1,xm-x);
1.126 brouard 1956: }
1957: } else {
1958: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1959: }
1960: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1961: fu=(*f)(u);
1962: if (fu <= fx) {
1963: if (u >= x) a=x; else b=x;
1964: SHFT(v,w,x,u)
1.183 brouard 1965: SHFT(fv,fw,fx,fu)
1966: } else {
1967: if (u < x) a=u; else b=u;
1968: if (fu <= fw || w == x) {
1.224 brouard 1969: v=w;
1970: w=u;
1971: fv=fw;
1972: fw=fu;
1.183 brouard 1973: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1974: v=u;
1975: fv=fu;
1.183 brouard 1976: }
1977: }
1.126 brouard 1978: }
1979: nrerror("Too many iterations in brent");
1980: *xmin=x;
1981: return fx;
1982: }
1983:
1984: /****************** mnbrak ***********************/
1985:
1986: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1987: double (*func)(double))
1.183 brouard 1988: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1989: the downhill direction (defined by the function as evaluated at the initial points) and returns
1990: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1991: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1992: */
1.126 brouard 1993: double ulim,u,r,q, dum;
1994: double fu;
1.187 brouard 1995:
1996: double scale=10.;
1997: int iterscale=0;
1998:
1999: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2000: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2001:
2002:
2003: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2004: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2005: /* *bx = *ax - (*ax - *bx)/scale; */
2006: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2007: /* } */
2008:
1.126 brouard 2009: if (*fb > *fa) {
2010: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2011: SHFT(dum,*fb,*fa,dum)
2012: }
1.126 brouard 2013: *cx=(*bx)+GOLD*(*bx-*ax);
2014: *fc=(*func)(*cx);
1.183 brouard 2015: #ifdef DEBUG
1.224 brouard 2016: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2017: 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 2018: #endif
1.224 brouard 2019: 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 2020: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2021: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2022: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2023: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2024: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2025: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2026: fu=(*func)(u);
1.163 brouard 2027: #ifdef DEBUG
2028: /* f(x)=A(x-u)**2+f(u) */
2029: double A, fparabu;
2030: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2031: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2032: 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);
2033: 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 2034: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2035: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2036: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2037: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2038: #endif
1.184 brouard 2039: #ifdef MNBRAKORIGINAL
1.183 brouard 2040: #else
1.191 brouard 2041: /* if (fu > *fc) { */
2042: /* #ifdef DEBUG */
2043: /* printf("mnbrak4 fu > fc \n"); */
2044: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2045: /* #endif */
2046: /* /\* 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 *\\/ *\/ */
2047: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2048: /* dum=u; /\* Shifting c and u *\/ */
2049: /* u = *cx; */
2050: /* *cx = dum; */
2051: /* dum = fu; */
2052: /* fu = *fc; */
2053: /* *fc =dum; */
2054: /* } else { /\* end *\/ */
2055: /* #ifdef DEBUG */
2056: /* printf("mnbrak3 fu < fc \n"); */
2057: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2058: /* #endif */
2059: /* dum=u; /\* Shifting c and u *\/ */
2060: /* u = *cx; */
2061: /* *cx = dum; */
2062: /* dum = fu; */
2063: /* fu = *fc; */
2064: /* *fc =dum; */
2065: /* } */
1.224 brouard 2066: #ifdef DEBUGMNBRAK
2067: double A, fparabu;
2068: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2069: fparabu= *fa - A*(*ax-u)*(*ax-u);
2070: 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);
2071: 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 2072: #endif
1.191 brouard 2073: dum=u; /* Shifting c and u */
2074: u = *cx;
2075: *cx = dum;
2076: dum = fu;
2077: fu = *fc;
2078: *fc =dum;
1.183 brouard 2079: #endif
1.162 brouard 2080: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2081: #ifdef DEBUG
1.224 brouard 2082: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2083: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2084: #endif
1.126 brouard 2085: fu=(*func)(u);
2086: if (fu < *fc) {
1.183 brouard 2087: #ifdef DEBUG
1.224 brouard 2088: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2089: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2090: #endif
2091: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2092: SHFT(*fb,*fc,fu,(*func)(u))
2093: #ifdef DEBUG
2094: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2095: #endif
2096: }
1.162 brouard 2097: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2098: #ifdef DEBUG
1.224 brouard 2099: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2100: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2101: #endif
1.126 brouard 2102: u=ulim;
2103: fu=(*func)(u);
1.183 brouard 2104: } else { /* u could be left to b (if r > q parabola has a maximum) */
2105: #ifdef DEBUG
1.224 brouard 2106: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2107: 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 2108: #endif
1.126 brouard 2109: u=(*cx)+GOLD*(*cx-*bx);
2110: fu=(*func)(u);
1.224 brouard 2111: #ifdef DEBUG
2112: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2113: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2114: #endif
1.183 brouard 2115: } /* end tests */
1.126 brouard 2116: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2117: SHFT(*fa,*fb,*fc,fu)
2118: #ifdef DEBUG
1.224 brouard 2119: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2120: 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 2121: #endif
2122: } /* 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 2123: }
2124:
2125: /*************** linmin ************************/
1.162 brouard 2126: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2127: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2128: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2129: the value of func at the returned location p . This is actually all accomplished by calling the
2130: routines mnbrak and brent .*/
1.126 brouard 2131: int ncom;
2132: double *pcom,*xicom;
2133: double (*nrfunc)(double []);
2134:
1.224 brouard 2135: #ifdef LINMINORIGINAL
1.126 brouard 2136: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2137: #else
2138: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2139: #endif
1.126 brouard 2140: {
2141: double brent(double ax, double bx, double cx,
2142: double (*f)(double), double tol, double *xmin);
2143: double f1dim(double x);
2144: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2145: double *fc, double (*func)(double));
2146: int j;
2147: double xx,xmin,bx,ax;
2148: double fx,fb,fa;
1.187 brouard 2149:
1.203 brouard 2150: #ifdef LINMINORIGINAL
2151: #else
2152: double scale=10., axs, xxs; /* Scale added for infinity */
2153: #endif
2154:
1.126 brouard 2155: ncom=n;
2156: pcom=vector(1,n);
2157: xicom=vector(1,n);
2158: nrfunc=func;
2159: for (j=1;j<=n;j++) {
2160: pcom[j]=p[j];
1.202 brouard 2161: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2162: }
1.187 brouard 2163:
1.203 brouard 2164: #ifdef LINMINORIGINAL
2165: xx=1.;
2166: #else
2167: axs=0.0;
2168: xxs=1.;
2169: do{
2170: xx= xxs;
2171: #endif
1.187 brouard 2172: ax=0.;
2173: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2174: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2175: /* 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)) */
2176: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2177: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2178: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2179: /* 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 2180: #ifdef LINMINORIGINAL
2181: #else
2182: if (fx != fx){
1.224 brouard 2183: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2184: printf("|");
2185: fprintf(ficlog,"|");
1.203 brouard 2186: #ifdef DEBUGLINMIN
1.224 brouard 2187: 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 2188: #endif
2189: }
1.224 brouard 2190: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2191: #endif
2192:
1.191 brouard 2193: #ifdef DEBUGLINMIN
2194: 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 2195: 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 2196: #endif
1.224 brouard 2197: #ifdef LINMINORIGINAL
2198: #else
2199: if(fb == fx){ /* Flat function in the direction */
2200: xmin=xx;
2201: *flat=1;
2202: }else{
2203: *flat=0;
2204: #endif
2205: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2206: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2207: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2208: /* fmin = f(p[j] + xmin * xi[j]) */
2209: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2210: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2211: #ifdef DEBUG
1.224 brouard 2212: 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);
2213: 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);
2214: #endif
2215: #ifdef LINMINORIGINAL
2216: #else
2217: }
1.126 brouard 2218: #endif
1.191 brouard 2219: #ifdef DEBUGLINMIN
2220: printf("linmin end ");
1.202 brouard 2221: fprintf(ficlog,"linmin end ");
1.191 brouard 2222: #endif
1.126 brouard 2223: for (j=1;j<=n;j++) {
1.203 brouard 2224: #ifdef LINMINORIGINAL
2225: xi[j] *= xmin;
2226: #else
2227: #ifdef DEBUGLINMIN
2228: if(xxs <1.0)
2229: printf(" before xi[%d]=%12.8f", j,xi[j]);
2230: #endif
2231: 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) */
2232: #ifdef DEBUGLINMIN
2233: if(xxs <1.0)
2234: 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 );
2235: #endif
2236: #endif
1.187 brouard 2237: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2238: }
1.191 brouard 2239: #ifdef DEBUGLINMIN
1.203 brouard 2240: printf("\n");
1.191 brouard 2241: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2242: 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 2243: for (j=1;j<=n;j++) {
1.202 brouard 2244: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2245: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2246: if(j % ncovmodel == 0){
1.191 brouard 2247: printf("\n");
1.202 brouard 2248: fprintf(ficlog,"\n");
2249: }
1.191 brouard 2250: }
1.203 brouard 2251: #else
1.191 brouard 2252: #endif
1.126 brouard 2253: free_vector(xicom,1,n);
2254: free_vector(pcom,1,n);
2255: }
2256:
2257:
2258: /*************** powell ************************/
1.162 brouard 2259: /*
2260: Minimization of a function func of n variables. Input consists of an initial starting point
2261: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2262: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2263: such that failure to decrease by more than this amount on one iteration signals doneness. On
2264: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2265: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2266: */
1.224 brouard 2267: #ifdef LINMINORIGINAL
2268: #else
2269: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2270: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2271: #endif
1.126 brouard 2272: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2273: double (*func)(double []))
2274: {
1.224 brouard 2275: #ifdef LINMINORIGINAL
2276: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2277: double (*func)(double []));
1.224 brouard 2278: #else
1.241 brouard 2279: void linmin(double p[], double xi[], int n, double *fret,
2280: double (*func)(double []),int *flat);
1.224 brouard 2281: #endif
1.239 brouard 2282: int i,ibig,j,jk,k;
1.126 brouard 2283: double del,t,*pt,*ptt,*xit;
1.181 brouard 2284: double directest;
1.126 brouard 2285: double fp,fptt;
2286: double *xits;
2287: int niterf, itmp;
1.224 brouard 2288: #ifdef LINMINORIGINAL
2289: #else
2290:
2291: flatdir=ivector(1,n);
2292: for (j=1;j<=n;j++) flatdir[j]=0;
2293: #endif
1.126 brouard 2294:
2295: pt=vector(1,n);
2296: ptt=vector(1,n);
2297: xit=vector(1,n);
2298: xits=vector(1,n);
2299: *fret=(*func)(p);
2300: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2301: rcurr_time = time(NULL);
1.126 brouard 2302: for (*iter=1;;++(*iter)) {
1.187 brouard 2303: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2304: ibig=0;
2305: del=0.0;
1.157 brouard 2306: rlast_time=rcurr_time;
2307: /* (void) gettimeofday(&curr_time,&tzp); */
2308: rcurr_time = time(NULL);
2309: curr_time = *localtime(&rcurr_time);
2310: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2311: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2312: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2313: for (i=1;i<=n;i++) {
1.126 brouard 2314: fprintf(ficrespow," %.12lf", p[i]);
2315: }
1.239 brouard 2316: fprintf(ficrespow,"\n");fflush(ficrespow);
2317: printf("\n#model= 1 + age ");
2318: fprintf(ficlog,"\n#model= 1 + age ");
2319: if(nagesqr==1){
1.241 brouard 2320: printf(" + age*age ");
2321: fprintf(ficlog," + age*age ");
1.239 brouard 2322: }
2323: for(j=1;j <=ncovmodel-2;j++){
2324: if(Typevar[j]==0) {
2325: printf(" + V%d ",Tvar[j]);
2326: fprintf(ficlog," + V%d ",Tvar[j]);
2327: }else if(Typevar[j]==1) {
2328: printf(" + V%d*age ",Tvar[j]);
2329: fprintf(ficlog," + V%d*age ",Tvar[j]);
2330: }else if(Typevar[j]==2) {
2331: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2332: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2333: }
2334: }
1.126 brouard 2335: printf("\n");
1.239 brouard 2336: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2337: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2338: fprintf(ficlog,"\n");
1.239 brouard 2339: for(i=1,jk=1; i <=nlstate; i++){
2340: for(k=1; k <=(nlstate+ndeath); k++){
2341: if (k != i) {
2342: printf("%d%d ",i,k);
2343: fprintf(ficlog,"%d%d ",i,k);
2344: for(j=1; j <=ncovmodel; j++){
2345: printf("%12.7f ",p[jk]);
2346: fprintf(ficlog,"%12.7f ",p[jk]);
2347: jk++;
2348: }
2349: printf("\n");
2350: fprintf(ficlog,"\n");
2351: }
2352: }
2353: }
1.241 brouard 2354: if(*iter <=3 && *iter >1){
1.157 brouard 2355: tml = *localtime(&rcurr_time);
2356: strcpy(strcurr,asctime(&tml));
2357: rforecast_time=rcurr_time;
1.126 brouard 2358: itmp = strlen(strcurr);
2359: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2360: strcurr[itmp-1]='\0';
1.162 brouard 2361: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2362: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2363: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2364: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2365: forecast_time = *localtime(&rforecast_time);
2366: strcpy(strfor,asctime(&forecast_time));
2367: itmp = strlen(strfor);
2368: if(strfor[itmp-1]=='\n')
2369: strfor[itmp-1]='\0';
2370: 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);
2371: 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 2372: }
2373: }
1.187 brouard 2374: for (i=1;i<=n;i++) { /* For each direction i */
2375: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2376: fptt=(*fret);
2377: #ifdef DEBUG
1.203 brouard 2378: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2379: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2380: #endif
1.203 brouard 2381: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2382: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2383: #ifdef LINMINORIGINAL
1.188 brouard 2384: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2385: #else
2386: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2387: flatdir[i]=flat; /* Function is vanishing in that direction i */
2388: #endif
2389: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2390: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2391: /* because that direction will be replaced unless the gain del is small */
2392: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2393: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2394: /* with the new direction. */
2395: del=fabs(fptt-(*fret));
2396: ibig=i;
1.126 brouard 2397: }
2398: #ifdef DEBUG
2399: printf("%d %.12e",i,(*fret));
2400: fprintf(ficlog,"%d %.12e",i,(*fret));
2401: for (j=1;j<=n;j++) {
1.224 brouard 2402: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2403: printf(" x(%d)=%.12e",j,xit[j]);
2404: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2405: }
2406: for(j=1;j<=n;j++) {
1.225 brouard 2407: printf(" p(%d)=%.12e",j,p[j]);
2408: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2409: }
2410: printf("\n");
2411: fprintf(ficlog,"\n");
2412: #endif
1.187 brouard 2413: } /* end loop on each direction i */
2414: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2415: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2416: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2417: for(j=1;j<=n;j++) {
1.302 ! brouard 2418: if(flatdir[j] >0){
! 2419: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
! 2420: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
! 2421: }
! 2422: /* printf("\n"); */
! 2423: /* fprintf(ficlog,"\n"); */
! 2424: }
1.243 brouard 2425: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2426: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2427: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2428: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2429: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2430: /* decreased of more than 3.84 */
2431: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2432: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2433: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2434:
1.188 brouard 2435: /* Starting the program with initial values given by a former maximization will simply change */
2436: /* the scales of the directions and the directions, because the are reset to canonical directions */
2437: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2438: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2439: #ifdef DEBUG
2440: int k[2],l;
2441: k[0]=1;
2442: k[1]=-1;
2443: printf("Max: %.12e",(*func)(p));
2444: fprintf(ficlog,"Max: %.12e",(*func)(p));
2445: for (j=1;j<=n;j++) {
2446: printf(" %.12e",p[j]);
2447: fprintf(ficlog," %.12e",p[j]);
2448: }
2449: printf("\n");
2450: fprintf(ficlog,"\n");
2451: for(l=0;l<=1;l++) {
2452: for (j=1;j<=n;j++) {
2453: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2454: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2455: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2456: }
2457: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2458: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2459: }
2460: #endif
2461:
1.224 brouard 2462: #ifdef LINMINORIGINAL
2463: #else
2464: free_ivector(flatdir,1,n);
2465: #endif
1.126 brouard 2466: free_vector(xit,1,n);
2467: free_vector(xits,1,n);
2468: free_vector(ptt,1,n);
2469: free_vector(pt,1,n);
2470: return;
1.192 brouard 2471: } /* enough precision */
1.240 brouard 2472: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2473: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2474: ptt[j]=2.0*p[j]-pt[j];
2475: xit[j]=p[j]-pt[j];
2476: pt[j]=p[j];
2477: }
1.181 brouard 2478: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2479: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2480: if (*iter <=4) {
1.225 brouard 2481: #else
2482: #endif
1.224 brouard 2483: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2484: #else
1.161 brouard 2485: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2486: #endif
1.162 brouard 2487: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2488: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2489: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2490: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2491: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2492: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2493: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2494: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2495: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2496: /* Even if f3 <f1, directest can be negative and t >0 */
2497: /* mu² and del² are equal when f3=f1 */
2498: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2499: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2500: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2501: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2502: #ifdef NRCORIGINAL
2503: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2504: #else
2505: 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 2506: t= t- del*SQR(fp-fptt);
1.183 brouard 2507: #endif
1.202 brouard 2508: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2509: #ifdef DEBUG
1.181 brouard 2510: 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);
2511: 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 2512: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2513: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2514: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2515: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2516: 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);
2517: 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);
2518: #endif
1.183 brouard 2519: #ifdef POWELLORIGINAL
2520: if (t < 0.0) { /* Then we use it for new direction */
2521: #else
1.182 brouard 2522: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2523: 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 2524: 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 2525: 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 2526: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2527: }
1.181 brouard 2528: if (directest < 0.0) { /* Then we use it for new direction */
2529: #endif
1.191 brouard 2530: #ifdef DEBUGLINMIN
1.234 brouard 2531: printf("Before linmin in direction P%d-P0\n",n);
2532: for (j=1;j<=n;j++) {
2533: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2534: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2535: if(j % ncovmodel == 0){
2536: printf("\n");
2537: fprintf(ficlog,"\n");
2538: }
2539: }
1.224 brouard 2540: #endif
2541: #ifdef LINMINORIGINAL
1.234 brouard 2542: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2543: #else
1.234 brouard 2544: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2545: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2546: #endif
1.234 brouard 2547:
1.191 brouard 2548: #ifdef DEBUGLINMIN
1.234 brouard 2549: for (j=1;j<=n;j++) {
2550: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2551: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2552: if(j % ncovmodel == 0){
2553: printf("\n");
2554: fprintf(ficlog,"\n");
2555: }
2556: }
1.224 brouard 2557: #endif
1.234 brouard 2558: for (j=1;j<=n;j++) {
2559: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2560: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2561: }
1.224 brouard 2562: #ifdef LINMINORIGINAL
2563: #else
1.234 brouard 2564: for (j=1, flatd=0;j<=n;j++) {
2565: if(flatdir[j]>0)
2566: flatd++;
2567: }
2568: if(flatd >0){
1.255 brouard 2569: printf("%d flat directions: ",flatd);
2570: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2571: for (j=1;j<=n;j++) {
2572: if(flatdir[j]>0){
2573: printf("%d ",j);
2574: fprintf(ficlog,"%d ",j);
2575: }
2576: }
2577: printf("\n");
2578: fprintf(ficlog,"\n");
2579: }
1.191 brouard 2580: #endif
1.234 brouard 2581: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2582: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2583:
1.126 brouard 2584: #ifdef DEBUG
1.234 brouard 2585: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2586: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2587: for(j=1;j<=n;j++){
2588: printf(" %lf",xit[j]);
2589: fprintf(ficlog," %lf",xit[j]);
2590: }
2591: printf("\n");
2592: fprintf(ficlog,"\n");
1.126 brouard 2593: #endif
1.192 brouard 2594: } /* end of t or directest negative */
1.224 brouard 2595: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2596: #else
1.234 brouard 2597: } /* end if (fptt < fp) */
1.192 brouard 2598: #endif
1.225 brouard 2599: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2600: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2601: #else
1.224 brouard 2602: #endif
1.234 brouard 2603: } /* loop iteration */
1.126 brouard 2604: }
1.234 brouard 2605:
1.126 brouard 2606: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2607:
1.235 brouard 2608: 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 2609: {
1.279 brouard 2610: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2611: * (and selected quantitative values in nres)
2612: * by left multiplying the unit
2613: * matrix by transitions matrix until convergence is reached with precision ftolpl
2614: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2615: * Wx is row vector: population in state 1, population in state 2, population dead
2616: * or prevalence in state 1, prevalence in state 2, 0
2617: * newm is the matrix after multiplications, its rows are identical at a factor.
2618: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2619: * Output is prlim.
2620: * Initial matrix pimij
2621: */
1.206 brouard 2622: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2623: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2624: /* 0, 0 , 1} */
2625: /*
2626: * and after some iteration: */
2627: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2628: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2629: /* 0, 0 , 1} */
2630: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2631: /* {0.51571254859325999, 0.4842874514067399, */
2632: /* 0.51326036147820708, 0.48673963852179264} */
2633: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2634:
1.126 brouard 2635: int i, ii,j,k;
1.209 brouard 2636: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2637: /* double **matprod2(); */ /* test */
1.218 brouard 2638: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2639: double **newm;
1.209 brouard 2640: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2641: int ncvloop=0;
1.288 brouard 2642: int first=0;
1.169 brouard 2643:
1.209 brouard 2644: min=vector(1,nlstate);
2645: max=vector(1,nlstate);
2646: meandiff=vector(1,nlstate);
2647:
1.218 brouard 2648: /* Starting with matrix unity */
1.126 brouard 2649: for (ii=1;ii<=nlstate+ndeath;ii++)
2650: for (j=1;j<=nlstate+ndeath;j++){
2651: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2652: }
1.169 brouard 2653:
2654: cov[1]=1.;
2655:
2656: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2657: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2658: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2659: ncvloop++;
1.126 brouard 2660: newm=savm;
2661: /* Covariates have to be included here again */
1.138 brouard 2662: cov[2]=agefin;
1.187 brouard 2663: if(nagesqr==1)
2664: cov[3]= agefin*agefin;;
1.234 brouard 2665: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2666: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2667: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2668: /* 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 2669: }
2670: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2671: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2672: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2673: /* 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 2674: }
1.237 brouard 2675: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2676: if(Dummy[Tvar[Tage[k]]]){
2677: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2678: } else{
1.235 brouard 2679: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2680: }
1.235 brouard 2681: /* 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 2682: }
1.237 brouard 2683: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2684: /* 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 2685: if(Dummy[Tvard[k][1]==0]){
2686: if(Dummy[Tvard[k][2]==0]){
2687: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2688: }else{
2689: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2690: }
2691: }else{
2692: if(Dummy[Tvard[k][2]==0]){
2693: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2694: }else{
2695: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2696: }
2697: }
1.234 brouard 2698: }
1.138 brouard 2699: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2700: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2701: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2702: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2703: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2704: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2705: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2706:
1.126 brouard 2707: savm=oldm;
2708: oldm=newm;
1.209 brouard 2709:
2710: for(j=1; j<=nlstate; j++){
2711: max[j]=0.;
2712: min[j]=1.;
2713: }
2714: for(i=1;i<=nlstate;i++){
2715: sumnew=0;
2716: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2717: for(j=1; j<=nlstate; j++){
2718: prlim[i][j]= newm[i][j]/(1-sumnew);
2719: max[j]=FMAX(max[j],prlim[i][j]);
2720: min[j]=FMIN(min[j],prlim[i][j]);
2721: }
2722: }
2723:
1.126 brouard 2724: maxmax=0.;
1.209 brouard 2725: for(j=1; j<=nlstate; j++){
2726: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2727: maxmax=FMAX(maxmax,meandiff[j]);
2728: /* 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 2729: } /* j loop */
1.203 brouard 2730: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2731: /* 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 2732: if(maxmax < ftolpl){
1.209 brouard 2733: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2734: free_vector(min,1,nlstate);
2735: free_vector(max,1,nlstate);
2736: free_vector(meandiff,1,nlstate);
1.126 brouard 2737: return prlim;
2738: }
1.288 brouard 2739: } /* agefin loop */
1.208 brouard 2740: /* After some age loop it doesn't converge */
1.288 brouard 2741: if(!first){
2742: first=1;
2743: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d). Others in log file only...\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
2744: }
2745: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
2746:
1.209 brouard 2747: /* 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); */
2748: free_vector(min,1,nlstate);
2749: free_vector(max,1,nlstate);
2750: free_vector(meandiff,1,nlstate);
1.208 brouard 2751:
1.169 brouard 2752: return prlim; /* should not reach here */
1.126 brouard 2753: }
2754:
1.217 brouard 2755:
2756: /**** Back Prevalence limit (stable or period prevalence) ****************/
2757:
1.218 brouard 2758: /* 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) */
2759: /* 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 2760: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2761: {
1.264 brouard 2762: /* 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 2763: matrix by transitions matrix until convergence is reached with precision ftolpl */
2764: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2765: /* Wx is row vector: population in state 1, population in state 2, population dead */
2766: /* or prevalence in state 1, prevalence in state 2, 0 */
2767: /* newm is the matrix after multiplications, its rows are identical at a factor */
2768: /* Initial matrix pimij */
2769: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2770: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2771: /* 0, 0 , 1} */
2772: /*
2773: * and after some iteration: */
2774: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2775: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2776: /* 0, 0 , 1} */
2777: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2778: /* {0.51571254859325999, 0.4842874514067399, */
2779: /* 0.51326036147820708, 0.48673963852179264} */
2780: /* If we start from prlim again, prlim tends to a constant matrix */
2781:
2782: int i, ii,j,k;
1.247 brouard 2783: int first=0;
1.217 brouard 2784: double *min, *max, *meandiff, maxmax,sumnew=0.;
2785: /* double **matprod2(); */ /* test */
2786: double **out, cov[NCOVMAX+1], **bmij();
2787: double **newm;
1.218 brouard 2788: double **dnewm, **doldm, **dsavm; /* for use */
2789: double **oldm, **savm; /* for use */
2790:
1.217 brouard 2791: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2792: int ncvloop=0;
2793:
2794: min=vector(1,nlstate);
2795: max=vector(1,nlstate);
2796: meandiff=vector(1,nlstate);
2797:
1.266 brouard 2798: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2799: oldm=oldms; savm=savms;
2800:
2801: /* Starting with matrix unity */
2802: for (ii=1;ii<=nlstate+ndeath;ii++)
2803: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2804: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2805: }
2806:
2807: cov[1]=1.;
2808:
2809: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2810: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2811: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2812: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2813: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2814: ncvloop++;
1.218 brouard 2815: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2816: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2817: /* Covariates have to be included here again */
2818: cov[2]=agefin;
2819: if(nagesqr==1)
2820: cov[3]= agefin*agefin;;
1.242 brouard 2821: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2822: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2823: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2824: /* 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 2825: }
2826: /* for (k=1; k<=cptcovn;k++) { */
2827: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2828: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2829: /* /\* 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])]); *\/ */
2830: /* } */
2831: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2832: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2833: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2834: /* 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]); */
2835: }
2836: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2837: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2838: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2839: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2840: for (k=1; k<=cptcovage;k++){ /* For product with age */
2841: if(Dummy[Tvar[Tage[k]]]){
2842: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2843: } else{
2844: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2845: }
2846: /* 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]); */
2847: }
2848: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2849: /* 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]); */
2850: if(Dummy[Tvard[k][1]==0]){
2851: if(Dummy[Tvard[k][2]==0]){
2852: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2853: }else{
2854: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2855: }
2856: }else{
2857: if(Dummy[Tvard[k][2]==0]){
2858: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2859: }else{
2860: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2861: }
2862: }
1.217 brouard 2863: }
2864:
2865: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2866: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2867: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2868: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2869: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2870: /* ij should be linked to the correct index of cov */
2871: /* age and covariate values ij are in 'cov', but we need to pass
2872: * ij for the observed prevalence at age and status and covariate
2873: * number: prevacurrent[(int)agefin][ii][ij]
2874: */
2875: /* 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 *\/ */
2876: /* 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 *\/ */
2877: 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 2878: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2879: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2880: /* for(i=1; i<=nlstate+ndeath; i++) { */
2881: /* printf("%d newm= ",i); */
2882: /* for(j=1;j<=nlstate+ndeath;j++) { */
2883: /* printf("%f ",newm[i][j]); */
2884: /* } */
2885: /* printf("oldm * "); */
2886: /* for(j=1;j<=nlstate+ndeath;j++) { */
2887: /* printf("%f ",oldm[i][j]); */
2888: /* } */
1.268 brouard 2889: /* printf(" bmmij "); */
1.266 brouard 2890: /* for(j=1;j<=nlstate+ndeath;j++) { */
2891: /* printf("%f ",pmmij[i][j]); */
2892: /* } */
2893: /* printf("\n"); */
2894: /* } */
2895: /* } */
1.217 brouard 2896: savm=oldm;
2897: oldm=newm;
1.266 brouard 2898:
1.217 brouard 2899: for(j=1; j<=nlstate; j++){
2900: max[j]=0.;
2901: min[j]=1.;
2902: }
2903: for(j=1; j<=nlstate; j++){
2904: for(i=1;i<=nlstate;i++){
1.234 brouard 2905: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2906: bprlim[i][j]= newm[i][j];
2907: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2908: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2909: }
2910: }
1.218 brouard 2911:
1.217 brouard 2912: maxmax=0.;
2913: for(i=1; i<=nlstate; i++){
2914: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2915: maxmax=FMAX(maxmax,meandiff[i]);
2916: /* 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 2917: } /* i loop */
1.217 brouard 2918: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2919: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2920: if(maxmax < ftolpl){
1.220 brouard 2921: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2922: free_vector(min,1,nlstate);
2923: free_vector(max,1,nlstate);
2924: free_vector(meandiff,1,nlstate);
2925: return bprlim;
2926: }
1.288 brouard 2927: } /* agefin loop */
1.217 brouard 2928: /* After some age loop it doesn't converge */
1.288 brouard 2929: if(!first){
1.247 brouard 2930: first=1;
2931: 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\
2932: 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);
2933: }
2934: 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 2935: 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);
2936: /* 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); */
2937: free_vector(min,1,nlstate);
2938: free_vector(max,1,nlstate);
2939: free_vector(meandiff,1,nlstate);
2940:
2941: return bprlim; /* should not reach here */
2942: }
2943:
1.126 brouard 2944: /*************** transition probabilities ***************/
2945:
2946: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2947: {
1.138 brouard 2948: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2949: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2950: model to the ncovmodel covariates (including constant and age).
2951: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2952: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2953: ncth covariate in the global vector x is given by the formula:
2954: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2955: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2956: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2957: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2958: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2959: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2960: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2961: */
2962: double s1, lnpijopii;
1.126 brouard 2963: /*double t34;*/
1.164 brouard 2964: int i,j, nc, ii, jj;
1.126 brouard 2965:
1.223 brouard 2966: for(i=1; i<= nlstate; i++){
2967: for(j=1; j<i;j++){
2968: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2969: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2970: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2971: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2972: }
2973: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2974: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2975: }
2976: for(j=i+1; j<=nlstate+ndeath;j++){
2977: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2978: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2979: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2980: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2981: }
2982: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2983: }
2984: }
1.218 brouard 2985:
1.223 brouard 2986: for(i=1; i<= nlstate; i++){
2987: s1=0;
2988: for(j=1; j<i; j++){
2989: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2990: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2991: }
2992: for(j=i+1; j<=nlstate+ndeath; j++){
2993: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2994: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2995: }
2996: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2997: ps[i][i]=1./(s1+1.);
2998: /* Computing other pijs */
2999: for(j=1; j<i; j++)
3000: ps[i][j]= exp(ps[i][j])*ps[i][i];
3001: for(j=i+1; j<=nlstate+ndeath; j++)
3002: ps[i][j]= exp(ps[i][j])*ps[i][i];
3003: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3004: } /* end i */
1.218 brouard 3005:
1.223 brouard 3006: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3007: for(jj=1; jj<= nlstate+ndeath; jj++){
3008: ps[ii][jj]=0;
3009: ps[ii][ii]=1;
3010: }
3011: }
1.294 brouard 3012:
3013:
1.223 brouard 3014: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3015: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3016: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3017: /* } */
3018: /* printf("\n "); */
3019: /* } */
3020: /* printf("\n ");printf("%lf ",cov[2]);*/
3021: /*
3022: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3023: goto end;*/
1.266 brouard 3024: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3025: }
3026:
1.218 brouard 3027: /*************** backward transition probabilities ***************/
3028:
3029: /* 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 ) */
3030: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3031: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3032: {
1.302 ! brouard 3033: /* Computes the backward probability at age agefin, cov[2], and covariate combination 'ij'. In fact cov is already filled and x too.
1.266 brouard 3034: * 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 3035: */
1.218 brouard 3036: int i, ii, j,k;
1.222 brouard 3037:
3038: double **out, **pmij();
3039: double sumnew=0.;
1.218 brouard 3040: double agefin;
1.292 brouard 3041: double k3=0.; /* constant of the w_x diagonal matrix (in order for B to sum to 1 even for death state) */
1.222 brouard 3042: double **dnewm, **dsavm, **doldm;
3043: double **bbmij;
3044:
1.218 brouard 3045: doldm=ddoldms; /* global pointers */
1.222 brouard 3046: dnewm=ddnewms;
3047: dsavm=ddsavms;
3048:
3049: agefin=cov[2];
1.268 brouard 3050: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3051: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3052: the observed prevalence (with this covariate ij) at beginning of transition */
3053: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3054:
3055: /* P_x */
1.266 brouard 3056: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3057: /* outputs pmmij which is a stochastic matrix in row */
3058:
3059: /* Diag(w_x) */
1.292 brouard 3060: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3061: sumnew=0.;
1.269 brouard 3062: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3063: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3064: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3065: sumnew+=prevacurrent[(int)agefin][ii][ij];
3066: }
3067: if(sumnew >0.01){ /* At least some value in the prevalence */
3068: for (ii=1;ii<=nlstate+ndeath;ii++){
3069: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3070: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3071: }
3072: }else{
3073: for (ii=1;ii<=nlstate+ndeath;ii++){
3074: for (j=1;j<=nlstate+ndeath;j++)
3075: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3076: }
3077: /* if(sumnew <0.9){ */
3078: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3079: /* } */
3080: }
3081: k3=0.0; /* We put the last diagonal to 0 */
3082: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3083: doldm[ii][ii]= k3;
3084: }
3085: /* End doldm, At the end doldm is diag[(w_i)] */
3086:
1.292 brouard 3087: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3088: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3089:
1.292 brouard 3090: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3091: /* 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 3092: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3093: sumnew=0.;
1.222 brouard 3094: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3095: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3096: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3097: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3098: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3099: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3100: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3101: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3102: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3103: /* }else */
1.268 brouard 3104: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3105: } /*End ii */
3106: } /* 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 */
3107:
1.292 brouard 3108: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3109: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3110: /* end bmij */
1.266 brouard 3111: return ps; /*pointer is unchanged */
1.218 brouard 3112: }
1.217 brouard 3113: /*************** transition probabilities ***************/
3114:
1.218 brouard 3115: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3116: {
3117: /* According to parameters values stored in x and the covariate's values stored in cov,
3118: computes the probability to be observed in state j being in state i by appying the
3119: model to the ncovmodel covariates (including constant and age).
3120: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3121: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3122: ncth covariate in the global vector x is given by the formula:
3123: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3124: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3125: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3126: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3127: Outputs ps[i][j] the probability to be observed in j being in j according to
3128: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3129: */
3130: double s1, lnpijopii;
3131: /*double t34;*/
3132: int i,j, nc, ii, jj;
3133:
1.234 brouard 3134: for(i=1; i<= nlstate; i++){
3135: for(j=1; j<i;j++){
3136: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3137: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3138: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3139: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3140: }
3141: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3142: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3143: }
3144: for(j=i+1; j<=nlstate+ndeath;j++){
3145: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3146: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3147: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3148: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3149: }
3150: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3151: }
3152: }
3153:
3154: for(i=1; i<= nlstate; i++){
3155: s1=0;
3156: for(j=1; j<i; j++){
3157: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3158: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3159: }
3160: for(j=i+1; j<=nlstate+ndeath; j++){
3161: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3162: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3163: }
3164: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3165: ps[i][i]=1./(s1+1.);
3166: /* Computing other pijs */
3167: for(j=1; j<i; j++)
3168: ps[i][j]= exp(ps[i][j])*ps[i][i];
3169: for(j=i+1; j<=nlstate+ndeath; j++)
3170: ps[i][j]= exp(ps[i][j])*ps[i][i];
3171: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3172: } /* end i */
3173:
3174: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3175: for(jj=1; jj<= nlstate+ndeath; jj++){
3176: ps[ii][jj]=0;
3177: ps[ii][ii]=1;
3178: }
3179: }
1.296 brouard 3180: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3181: for(jj=1; jj<= nlstate+ndeath; jj++){
3182: s1=0.;
3183: for(ii=1; ii<= nlstate+ndeath; ii++){
3184: s1+=ps[ii][jj];
3185: }
3186: for(ii=1; ii<= nlstate; ii++){
3187: ps[ii][jj]=ps[ii][jj]/s1;
3188: }
3189: }
3190: /* Transposition */
3191: for(jj=1; jj<= nlstate+ndeath; jj++){
3192: for(ii=jj; ii<= nlstate+ndeath; ii++){
3193: s1=ps[ii][jj];
3194: ps[ii][jj]=ps[jj][ii];
3195: ps[jj][ii]=s1;
3196: }
3197: }
3198: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3199: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3200: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3201: /* } */
3202: /* printf("\n "); */
3203: /* } */
3204: /* printf("\n ");printf("%lf ",cov[2]);*/
3205: /*
3206: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3207: goto end;*/
3208: return ps;
1.217 brouard 3209: }
3210:
3211:
1.126 brouard 3212: /**************** Product of 2 matrices ******************/
3213:
1.145 brouard 3214: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3215: {
3216: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3217: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3218: /* in, b, out are matrice of pointers which should have been initialized
3219: before: only the contents of out is modified. The function returns
3220: a pointer to pointers identical to out */
1.145 brouard 3221: int i, j, k;
1.126 brouard 3222: for(i=nrl; i<= nrh; i++)
1.145 brouard 3223: for(k=ncolol; k<=ncoloh; k++){
3224: out[i][k]=0.;
3225: for(j=ncl; j<=nch; j++)
3226: out[i][k] +=in[i][j]*b[j][k];
3227: }
1.126 brouard 3228: return out;
3229: }
3230:
3231:
3232: /************* Higher Matrix Product ***************/
3233:
1.235 brouard 3234: 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 3235: {
1.218 brouard 3236: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3237: 'nhstepm*hstepm*stepm' months (i.e. until
3238: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3239: nhstepm*hstepm matrices.
3240: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3241: (typically every 2 years instead of every month which is too big
3242: for the memory).
3243: Model is determined by parameters x and covariates have to be
3244: included manually here.
3245:
3246: */
3247:
3248: int i, j, d, h, k;
1.131 brouard 3249: double **out, cov[NCOVMAX+1];
1.126 brouard 3250: double **newm;
1.187 brouard 3251: double agexact;
1.214 brouard 3252: double agebegin, ageend;
1.126 brouard 3253:
3254: /* Hstepm could be zero and should return the unit matrix */
3255: for (i=1;i<=nlstate+ndeath;i++)
3256: for (j=1;j<=nlstate+ndeath;j++){
3257: oldm[i][j]=(i==j ? 1.0 : 0.0);
3258: po[i][j][0]=(i==j ? 1.0 : 0.0);
3259: }
3260: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3261: for(h=1; h <=nhstepm; h++){
3262: for(d=1; d <=hstepm; d++){
3263: newm=savm;
3264: /* Covariates have to be included here again */
3265: cov[1]=1.;
1.214 brouard 3266: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3267: cov[2]=agexact;
3268: if(nagesqr==1)
1.227 brouard 3269: cov[3]= agexact*agexact;
1.235 brouard 3270: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3271: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3272: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3273: /* 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)); */
3274: }
3275: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3276: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3277: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3278: /* 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]); */
3279: }
3280: for (k=1; k<=cptcovage;k++){
3281: if(Dummy[Tvar[Tage[k]]]){
3282: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3283: } else{
3284: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3285: }
3286: /* 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]); */
3287: }
3288: for (k=1; k<=cptcovprod;k++){ /* */
3289: /* 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]); */
3290: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3291: }
3292: /* for (k=1; k<=cptcovn;k++) */
3293: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3294: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3295: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3296: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3297: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3298:
3299:
1.126 brouard 3300: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3301: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3302: /* right multiplication of oldm by the current matrix */
1.126 brouard 3303: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3304: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3305: /* if((int)age == 70){ */
3306: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3307: /* for(i=1; i<=nlstate+ndeath; i++) { */
3308: /* printf("%d pmmij ",i); */
3309: /* for(j=1;j<=nlstate+ndeath;j++) { */
3310: /* printf("%f ",pmmij[i][j]); */
3311: /* } */
3312: /* printf(" oldm "); */
3313: /* for(j=1;j<=nlstate+ndeath;j++) { */
3314: /* printf("%f ",oldm[i][j]); */
3315: /* } */
3316: /* printf("\n"); */
3317: /* } */
3318: /* } */
1.126 brouard 3319: savm=oldm;
3320: oldm=newm;
3321: }
3322: for(i=1; i<=nlstate+ndeath; i++)
3323: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3324: po[i][j][h]=newm[i][j];
3325: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3326: }
1.128 brouard 3327: /*printf("h=%d ",h);*/
1.126 brouard 3328: } /* end h */
1.267 brouard 3329: /* printf("\n H=%d \n",h); */
1.126 brouard 3330: return po;
3331: }
3332:
1.217 brouard 3333: /************* Higher Back Matrix Product ***************/
1.218 brouard 3334: /* 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 3335: 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 3336: {
1.266 brouard 3337: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3338: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3339: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3340: nhstepm*hstepm matrices.
3341: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3342: (typically every 2 years instead of every month which is too big
1.217 brouard 3343: for the memory).
1.218 brouard 3344: Model is determined by parameters x and covariates have to be
1.266 brouard 3345: included manually here. Then we use a call to bmij(x and cov)
3346: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3347: */
1.217 brouard 3348:
3349: int i, j, d, h, k;
1.266 brouard 3350: double **out, cov[NCOVMAX+1], **bmij();
3351: double **newm, ***newmm;
1.217 brouard 3352: double agexact;
3353: double agebegin, ageend;
1.222 brouard 3354: double **oldm, **savm;
1.217 brouard 3355:
1.266 brouard 3356: newmm=po; /* To be saved */
3357: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3358: /* Hstepm could be zero and should return the unit matrix */
3359: for (i=1;i<=nlstate+ndeath;i++)
3360: for (j=1;j<=nlstate+ndeath;j++){
3361: oldm[i][j]=(i==j ? 1.0 : 0.0);
3362: po[i][j][0]=(i==j ? 1.0 : 0.0);
3363: }
3364: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3365: for(h=1; h <=nhstepm; h++){
3366: for(d=1; d <=hstepm; d++){
3367: newm=savm;
3368: /* Covariates have to be included here again */
3369: cov[1]=1.;
1.271 brouard 3370: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3371: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3372: cov[2]=agexact;
3373: if(nagesqr==1)
1.222 brouard 3374: cov[3]= agexact*agexact;
1.266 brouard 3375: for (k=1; k<=cptcovn;k++){
3376: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3377: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3378: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3379: /* 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)); */
3380: }
1.267 brouard 3381: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3382: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3383: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3384: /* 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]); */
3385: }
3386: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3387: if(Dummy[Tvar[Tage[k]]]){
3388: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3389: } else{
3390: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3391: }
3392: /* 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]); */
3393: }
3394: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3395: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3396: }
1.217 brouard 3397: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3398: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3399:
1.218 brouard 3400: /* Careful transposed matrix */
1.266 brouard 3401: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3402: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3403: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3404: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3405: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3406: /* if((int)age == 70){ */
3407: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3408: /* for(i=1; i<=nlstate+ndeath; i++) { */
3409: /* printf("%d pmmij ",i); */
3410: /* for(j=1;j<=nlstate+ndeath;j++) { */
3411: /* printf("%f ",pmmij[i][j]); */
3412: /* } */
3413: /* printf(" oldm "); */
3414: /* for(j=1;j<=nlstate+ndeath;j++) { */
3415: /* printf("%f ",oldm[i][j]); */
3416: /* } */
3417: /* printf("\n"); */
3418: /* } */
3419: /* } */
3420: savm=oldm;
3421: oldm=newm;
3422: }
3423: for(i=1; i<=nlstate+ndeath; i++)
3424: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3425: po[i][j][h]=newm[i][j];
1.268 brouard 3426: /* if(h==nhstepm) */
3427: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3428: }
1.268 brouard 3429: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3430: } /* end h */
1.268 brouard 3431: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3432: return po;
3433: }
3434:
3435:
1.162 brouard 3436: #ifdef NLOPT
3437: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3438: double fret;
3439: double *xt;
3440: int j;
3441: myfunc_data *d2 = (myfunc_data *) pd;
3442: /* xt = (p1-1); */
3443: xt=vector(1,n);
3444: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3445:
3446: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3447: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3448: printf("Function = %.12lf ",fret);
3449: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3450: printf("\n");
3451: free_vector(xt,1,n);
3452: return fret;
3453: }
3454: #endif
1.126 brouard 3455:
3456: /*************** log-likelihood *************/
3457: double func( double *x)
3458: {
1.226 brouard 3459: int i, ii, j, k, mi, d, kk;
3460: int ioffset=0;
3461: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3462: double **out;
3463: double lli; /* Individual log likelihood */
3464: int s1, s2;
1.228 brouard 3465: 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 3466: double bbh, survp;
3467: long ipmx;
3468: double agexact;
3469: /*extern weight */
3470: /* We are differentiating ll according to initial status */
3471: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3472: /*for(i=1;i<imx;i++)
3473: printf(" %d\n",s[4][i]);
3474: */
1.162 brouard 3475:
1.226 brouard 3476: ++countcallfunc;
1.162 brouard 3477:
1.226 brouard 3478: cov[1]=1.;
1.126 brouard 3479:
1.226 brouard 3480: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3481: ioffset=0;
1.226 brouard 3482: if(mle==1){
3483: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3484: /* Computes the values of the ncovmodel covariates of the model
3485: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3486: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3487: to be observed in j being in i according to the model.
3488: */
1.243 brouard 3489: ioffset=2+nagesqr ;
1.233 brouard 3490: /* Fixed */
1.234 brouard 3491: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3492: 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)*/
3493: }
1.226 brouard 3494: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3495: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3496: has been calculated etc */
3497: /* For an individual i, wav[i] gives the number of effective waves */
3498: /* We compute the contribution to Likelihood of each effective transition
3499: mw[mi][i] is real wave of the mi th effectve wave */
3500: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3501: s2=s[mw[mi+1][i]][i];
3502: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3503: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3504: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3505: */
3506: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3507: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3508: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3509: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3510: }
3511: for (ii=1;ii<=nlstate+ndeath;ii++)
3512: for (j=1;j<=nlstate+ndeath;j++){
3513: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3514: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3515: }
3516: for(d=0; d<dh[mi][i]; d++){
3517: newm=savm;
3518: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3519: cov[2]=agexact;
3520: if(nagesqr==1)
3521: cov[3]= agexact*agexact; /* Should be changed here */
3522: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3523: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3524: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3525: else
3526: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3527: }
3528: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3529: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3530: savm=oldm;
3531: oldm=newm;
3532: } /* end mult */
3533:
3534: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3535: /* But now since version 0.9 we anticipate for bias at large stepm.
3536: * If stepm is larger than one month (smallest stepm) and if the exact delay
3537: * (in months) between two waves is not a multiple of stepm, we rounded to
3538: * the nearest (and in case of equal distance, to the lowest) interval but now
3539: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3540: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3541: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3542: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3543: * -stepm/2 to stepm/2 .
3544: * For stepm=1 the results are the same as for previous versions of Imach.
3545: * For stepm > 1 the results are less biased than in previous versions.
3546: */
1.234 brouard 3547: s1=s[mw[mi][i]][i];
3548: s2=s[mw[mi+1][i]][i];
3549: bbh=(double)bh[mi][i]/(double)stepm;
3550: /* bias bh is positive if real duration
3551: * is higher than the multiple of stepm and negative otherwise.
3552: */
3553: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3554: if( s2 > nlstate){
3555: /* i.e. if s2 is a death state and if the date of death is known
3556: then the contribution to the likelihood is the probability to
3557: die between last step unit time and current step unit time,
3558: which is also equal to probability to die before dh
3559: minus probability to die before dh-stepm .
3560: In version up to 0.92 likelihood was computed
3561: as if date of death was unknown. Death was treated as any other
3562: health state: the date of the interview describes the actual state
3563: and not the date of a change in health state. The former idea was
3564: to consider that at each interview the state was recorded
3565: (healthy, disable or death) and IMaCh was corrected; but when we
3566: introduced the exact date of death then we should have modified
3567: the contribution of an exact death to the likelihood. This new
3568: contribution is smaller and very dependent of the step unit
3569: stepm. It is no more the probability to die between last interview
3570: and month of death but the probability to survive from last
3571: interview up to one month before death multiplied by the
3572: probability to die within a month. Thanks to Chris
3573: Jackson for correcting this bug. Former versions increased
3574: mortality artificially. The bad side is that we add another loop
3575: which slows down the processing. The difference can be up to 10%
3576: lower mortality.
3577: */
3578: /* If, at the beginning of the maximization mostly, the
3579: cumulative probability or probability to be dead is
3580: constant (ie = 1) over time d, the difference is equal to
3581: 0. out[s1][3] = savm[s1][3]: probability, being at state
3582: s1 at precedent wave, to be dead a month before current
3583: wave is equal to probability, being at state s1 at
3584: precedent wave, to be dead at mont of the current
3585: wave. Then the observed probability (that this person died)
3586: is null according to current estimated parameter. In fact,
3587: it should be very low but not zero otherwise the log go to
3588: infinity.
3589: */
1.183 brouard 3590: /* #ifdef INFINITYORIGINAL */
3591: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3592: /* #else */
3593: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3594: /* lli=log(mytinydouble); */
3595: /* else */
3596: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3597: /* #endif */
1.226 brouard 3598: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3599:
1.226 brouard 3600: } else if ( s2==-1 ) { /* alive */
3601: for (j=1,survp=0. ; j<=nlstate; j++)
3602: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3603: /*survp += out[s1][j]; */
3604: lli= log(survp);
3605: }
3606: else if (s2==-4) {
3607: for (j=3,survp=0. ; j<=nlstate; j++)
3608: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3609: lli= log(survp);
3610: }
3611: else if (s2==-5) {
3612: for (j=1,survp=0. ; j<=2; j++)
3613: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3614: lli= log(survp);
3615: }
3616: else{
3617: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3618: /* 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 */
3619: }
3620: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3621: /*if(lli ==000.0)*/
3622: /*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); */
3623: ipmx +=1;
3624: sw += weight[i];
3625: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3626: /* if (lli < log(mytinydouble)){ */
3627: /* 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); */
3628: /* 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]); */
3629: /* } */
3630: } /* end of wave */
3631: } /* end of individual */
3632: } else if(mle==2){
3633: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3634: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3635: for(mi=1; mi<= wav[i]-1; mi++){
3636: for (ii=1;ii<=nlstate+ndeath;ii++)
3637: for (j=1;j<=nlstate+ndeath;j++){
3638: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3639: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3640: }
3641: for(d=0; d<=dh[mi][i]; d++){
3642: newm=savm;
3643: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3644: cov[2]=agexact;
3645: if(nagesqr==1)
3646: cov[3]= agexact*agexact;
3647: for (kk=1; kk<=cptcovage;kk++) {
3648: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3649: }
3650: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3651: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3652: savm=oldm;
3653: oldm=newm;
3654: } /* end mult */
3655:
3656: s1=s[mw[mi][i]][i];
3657: s2=s[mw[mi+1][i]][i];
3658: bbh=(double)bh[mi][i]/(double)stepm;
3659: 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 */
3660: ipmx +=1;
3661: sw += weight[i];
3662: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3663: } /* end of wave */
3664: } /* end of individual */
3665: } else if(mle==3){ /* exponential inter-extrapolation */
3666: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3667: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3668: for(mi=1; mi<= wav[i]-1; mi++){
3669: for (ii=1;ii<=nlstate+ndeath;ii++)
3670: for (j=1;j<=nlstate+ndeath;j++){
3671: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3672: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3673: }
3674: for(d=0; d<dh[mi][i]; d++){
3675: newm=savm;
3676: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3677: cov[2]=agexact;
3678: if(nagesqr==1)
3679: cov[3]= agexact*agexact;
3680: for (kk=1; kk<=cptcovage;kk++) {
3681: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3682: }
3683: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3684: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3685: savm=oldm;
3686: oldm=newm;
3687: } /* end mult */
3688:
3689: s1=s[mw[mi][i]][i];
3690: s2=s[mw[mi+1][i]][i];
3691: bbh=(double)bh[mi][i]/(double)stepm;
3692: 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 */
3693: ipmx +=1;
3694: sw += weight[i];
3695: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3696: } /* end of wave */
3697: } /* end of individual */
3698: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3699: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3700: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3701: for(mi=1; mi<= wav[i]-1; mi++){
3702: for (ii=1;ii<=nlstate+ndeath;ii++)
3703: for (j=1;j<=nlstate+ndeath;j++){
3704: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3705: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3706: }
3707: for(d=0; d<dh[mi][i]; d++){
3708: newm=savm;
3709: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3710: cov[2]=agexact;
3711: if(nagesqr==1)
3712: cov[3]= agexact*agexact;
3713: for (kk=1; kk<=cptcovage;kk++) {
3714: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3715: }
1.126 brouard 3716:
1.226 brouard 3717: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3718: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3719: savm=oldm;
3720: oldm=newm;
3721: } /* end mult */
3722:
3723: s1=s[mw[mi][i]][i];
3724: s2=s[mw[mi+1][i]][i];
3725: if( s2 > nlstate){
3726: lli=log(out[s1][s2] - savm[s1][s2]);
3727: } else if ( s2==-1 ) { /* alive */
3728: for (j=1,survp=0. ; j<=nlstate; j++)
3729: survp += out[s1][j];
3730: lli= log(survp);
3731: }else{
3732: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3733: }
3734: ipmx +=1;
3735: sw += weight[i];
3736: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3737: /* 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 3738: } /* end of wave */
3739: } /* end of individual */
3740: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3741: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3742: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3743: for(mi=1; mi<= wav[i]-1; mi++){
3744: for (ii=1;ii<=nlstate+ndeath;ii++)
3745: for (j=1;j<=nlstate+ndeath;j++){
3746: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3747: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3748: }
3749: for(d=0; d<dh[mi][i]; d++){
3750: newm=savm;
3751: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3752: cov[2]=agexact;
3753: if(nagesqr==1)
3754: cov[3]= agexact*agexact;
3755: for (kk=1; kk<=cptcovage;kk++) {
3756: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3757: }
1.126 brouard 3758:
1.226 brouard 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;
3763: } /* end mult */
3764:
3765: s1=s[mw[mi][i]][i];
3766: s2=s[mw[mi+1][i]][i];
3767: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3768: ipmx +=1;
3769: sw += weight[i];
3770: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3771: /*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]);*/
3772: } /* end of wave */
3773: } /* end of individual */
3774: } /* End of if */
3775: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3776: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3777: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3778: return -l;
1.126 brouard 3779: }
3780:
3781: /*************** log-likelihood *************/
3782: double funcone( double *x)
3783: {
1.228 brouard 3784: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3785: int i, ii, j, k, mi, d, kk;
1.228 brouard 3786: int ioffset=0;
1.131 brouard 3787: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3788: double **out;
3789: double lli; /* Individual log likelihood */
3790: double llt;
3791: int s1, s2;
1.228 brouard 3792: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3793:
1.126 brouard 3794: double bbh, survp;
1.187 brouard 3795: double agexact;
1.214 brouard 3796: double agebegin, ageend;
1.126 brouard 3797: /*extern weight */
3798: /* We are differentiating ll according to initial status */
3799: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3800: /*for(i=1;i<imx;i++)
3801: printf(" %d\n",s[4][i]);
3802: */
3803: cov[1]=1.;
3804:
3805: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3806: ioffset=0;
3807: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3808: /* ioffset=2+nagesqr+cptcovage; */
3809: ioffset=2+nagesqr;
1.232 brouard 3810: /* Fixed */
1.224 brouard 3811: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3812: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3813: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3814: 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)*/
3815: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3816: /* cov[2+6]=covar[Tvar[6]][i]; */
3817: /* cov[2+6]=covar[2][i]; V2 */
3818: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3819: /* cov[2+7]=covar[Tvar[7]][i]; */
3820: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3821: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3822: /* cov[2+9]=covar[Tvar[9]][i]; */
3823: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3824: }
1.232 brouard 3825: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3826: /* 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?)*\/ */
3827: /* } */
1.231 brouard 3828: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3829: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3830: /* } */
1.225 brouard 3831:
1.233 brouard 3832:
3833: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3834: /* Wave varying (but not age varying) */
3835: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3836: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3837: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3838: }
1.232 brouard 3839: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3840: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3841: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3842: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3843: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3844: /* 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 3845: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3846: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3847: /* /\* 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]); *\/ */
3848: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3849: /* } */
1.126 brouard 3850: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3851: for (j=1;j<=nlstate+ndeath;j++){
3852: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3853: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3854: }
1.214 brouard 3855:
3856: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3857: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3858: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3859: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3860: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3861: and mw[mi+1][i]. dh depends on stepm.*/
3862: newm=savm;
1.247 brouard 3863: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3864: cov[2]=agexact;
3865: if(nagesqr==1)
3866: cov[3]= agexact*agexact;
3867: for (kk=1; kk<=cptcovage;kk++) {
3868: if(!FixedV[Tvar[Tage[kk]]])
3869: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3870: else
3871: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3872: }
3873: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3874: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3875: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3876: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3877: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3878: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3879: savm=oldm;
3880: oldm=newm;
1.126 brouard 3881: } /* end mult */
3882:
3883: s1=s[mw[mi][i]][i];
3884: s2=s[mw[mi+1][i]][i];
1.217 brouard 3885: /* if(s2==-1){ */
1.268 brouard 3886: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3887: /* /\* exit(1); *\/ */
3888: /* } */
1.126 brouard 3889: bbh=(double)bh[mi][i]/(double)stepm;
3890: /* bias is positive if real duration
3891: * is higher than the multiple of stepm and negative otherwise.
3892: */
3893: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3894: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3895: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3896: for (j=1,survp=0. ; j<=nlstate; j++)
3897: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3898: lli= log(survp);
1.126 brouard 3899: }else if (mle==1){
1.242 brouard 3900: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3901: } else if(mle==2){
1.242 brouard 3902: 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 3903: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3904: 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 3905: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3906: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3907: } else{ /* mle=0 back to 1 */
1.242 brouard 3908: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3909: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3910: } /* End of if */
3911: ipmx +=1;
3912: sw += weight[i];
3913: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3914: /*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 3915: if(globpr){
1.246 brouard 3916: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3917: %11.6f %11.6f %11.6f ", \
1.242 brouard 3918: 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 3919: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3920: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3921: llt +=ll[k]*gipmx/gsw;
3922: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3923: }
3924: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3925: }
1.232 brouard 3926: } /* end of wave */
3927: } /* end of individual */
3928: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3929: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3930: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3931: if(globpr==0){ /* First time we count the contributions and weights */
3932: gipmx=ipmx;
3933: gsw=sw;
3934: }
3935: return -l;
1.126 brouard 3936: }
3937:
3938:
3939: /*************** function likelione ***********/
1.292 brouard 3940: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3941: {
3942: /* This routine should help understanding what is done with
3943: the selection of individuals/waves and
3944: to check the exact contribution to the likelihood.
3945: Plotting could be done.
3946: */
3947: int k;
3948:
3949: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3950: strcpy(fileresilk,"ILK_");
1.202 brouard 3951: strcat(fileresilk,fileresu);
1.126 brouard 3952: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3953: printf("Problem with resultfile: %s\n", fileresilk);
3954: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3955: }
1.214 brouard 3956: 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");
3957: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3958: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3959: for(k=1; k<=nlstate; k++)
3960: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3961: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3962: }
3963:
1.292 brouard 3964: *fretone=(*func)(p);
1.126 brouard 3965: if(*globpri !=0){
3966: fclose(ficresilk);
1.205 brouard 3967: if (mle ==0)
3968: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3969: else if(mle >=1)
3970: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3971: 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));
1.274 brouard 3972: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3973:
3974: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3975: 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 3976: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3977: }
1.207 brouard 3978: 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 3979: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3980: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3981: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3982: fflush(fichtm);
1.205 brouard 3983: }
1.126 brouard 3984: return;
3985: }
3986:
3987:
3988: /*********** Maximum Likelihood Estimation ***************/
3989:
3990: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3991: {
1.165 brouard 3992: int i,j, iter=0;
1.126 brouard 3993: double **xi;
3994: double fret;
3995: double fretone; /* Only one call to likelihood */
3996: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3997:
3998: #ifdef NLOPT
3999: int creturn;
4000: nlopt_opt opt;
4001: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4002: double *lb;
4003: double minf; /* the minimum objective value, upon return */
4004: double * p1; /* Shifted parameters from 0 instead of 1 */
4005: myfunc_data dinst, *d = &dinst;
4006: #endif
4007:
4008:
1.126 brouard 4009: xi=matrix(1,npar,1,npar);
4010: for (i=1;i<=npar;i++)
4011: for (j=1;j<=npar;j++)
4012: xi[i][j]=(i==j ? 1.0 : 0.0);
4013: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4014: strcpy(filerespow,"POW_");
1.126 brouard 4015: strcat(filerespow,fileres);
4016: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4017: printf("Problem with resultfile: %s\n", filerespow);
4018: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4019: }
4020: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4021: for (i=1;i<=nlstate;i++)
4022: for(j=1;j<=nlstate+ndeath;j++)
4023: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4024: fprintf(ficrespow,"\n");
1.162 brouard 4025: #ifdef POWELL
1.126 brouard 4026: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4027: #endif
1.126 brouard 4028:
1.162 brouard 4029: #ifdef NLOPT
4030: #ifdef NEWUOA
4031: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4032: #else
4033: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4034: #endif
4035: lb=vector(0,npar-1);
4036: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4037: nlopt_set_lower_bounds(opt, lb);
4038: nlopt_set_initial_step1(opt, 0.1);
4039:
4040: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4041: d->function = func;
4042: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4043: nlopt_set_min_objective(opt, myfunc, d);
4044: nlopt_set_xtol_rel(opt, ftol);
4045: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4046: printf("nlopt failed! %d\n",creturn);
4047: }
4048: else {
4049: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4050: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4051: iter=1; /* not equal */
4052: }
4053: nlopt_destroy(opt);
4054: #endif
1.126 brouard 4055: free_matrix(xi,1,npar,1,npar);
4056: fclose(ficrespow);
1.203 brouard 4057: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4058: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4059: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4060:
4061: }
4062:
4063: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4064: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4065: {
4066: double **a,**y,*x,pd;
1.203 brouard 4067: /* double **hess; */
1.164 brouard 4068: int i, j;
1.126 brouard 4069: int *indx;
4070:
4071: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4072: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4073: void lubksb(double **a, int npar, int *indx, double b[]) ;
4074: void ludcmp(double **a, int npar, int *indx, double *d) ;
4075: double gompertz(double p[]);
1.203 brouard 4076: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4077:
4078: printf("\nCalculation of the hessian matrix. Wait...\n");
4079: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4080: for (i=1;i<=npar;i++){
1.203 brouard 4081: printf("%d-",i);fflush(stdout);
4082: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4083:
4084: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4085:
4086: /* printf(" %f ",p[i]);
4087: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4088: }
4089:
4090: for (i=1;i<=npar;i++) {
4091: for (j=1;j<=npar;j++) {
4092: if (j>i) {
1.203 brouard 4093: printf(".%d-%d",i,j);fflush(stdout);
4094: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4095: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4096:
4097: hess[j][i]=hess[i][j];
4098: /*printf(" %lf ",hess[i][j]);*/
4099: }
4100: }
4101: }
4102: printf("\n");
4103: fprintf(ficlog,"\n");
4104:
4105: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4106: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4107:
4108: a=matrix(1,npar,1,npar);
4109: y=matrix(1,npar,1,npar);
4110: x=vector(1,npar);
4111: indx=ivector(1,npar);
4112: for (i=1;i<=npar;i++)
4113: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4114: ludcmp(a,npar,indx,&pd);
4115:
4116: for (j=1;j<=npar;j++) {
4117: for (i=1;i<=npar;i++) x[i]=0;
4118: x[j]=1;
4119: lubksb(a,npar,indx,x);
4120: for (i=1;i<=npar;i++){
4121: matcov[i][j]=x[i];
4122: }
4123: }
4124:
4125: printf("\n#Hessian matrix#\n");
4126: fprintf(ficlog,"\n#Hessian matrix#\n");
4127: for (i=1;i<=npar;i++) {
4128: for (j=1;j<=npar;j++) {
1.203 brouard 4129: printf("%.6e ",hess[i][j]);
4130: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4131: }
4132: printf("\n");
4133: fprintf(ficlog,"\n");
4134: }
4135:
1.203 brouard 4136: /* printf("\n#Covariance matrix#\n"); */
4137: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4138: /* for (i=1;i<=npar;i++) { */
4139: /* for (j=1;j<=npar;j++) { */
4140: /* printf("%.6e ",matcov[i][j]); */
4141: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4142: /* } */
4143: /* printf("\n"); */
4144: /* fprintf(ficlog,"\n"); */
4145: /* } */
4146:
1.126 brouard 4147: /* Recompute Inverse */
1.203 brouard 4148: /* for (i=1;i<=npar;i++) */
4149: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4150: /* ludcmp(a,npar,indx,&pd); */
4151:
4152: /* printf("\n#Hessian matrix recomputed#\n"); */
4153:
4154: /* for (j=1;j<=npar;j++) { */
4155: /* for (i=1;i<=npar;i++) x[i]=0; */
4156: /* x[j]=1; */
4157: /* lubksb(a,npar,indx,x); */
4158: /* for (i=1;i<=npar;i++){ */
4159: /* y[i][j]=x[i]; */
4160: /* printf("%.3e ",y[i][j]); */
4161: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4162: /* } */
4163: /* printf("\n"); */
4164: /* fprintf(ficlog,"\n"); */
4165: /* } */
4166:
4167: /* Verifying the inverse matrix */
4168: #ifdef DEBUGHESS
4169: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4170:
1.203 brouard 4171: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4172: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4173:
4174: for (j=1;j<=npar;j++) {
4175: for (i=1;i<=npar;i++){
1.203 brouard 4176: printf("%.2f ",y[i][j]);
4177: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4178: }
4179: printf("\n");
4180: fprintf(ficlog,"\n");
4181: }
1.203 brouard 4182: #endif
1.126 brouard 4183:
4184: free_matrix(a,1,npar,1,npar);
4185: free_matrix(y,1,npar,1,npar);
4186: free_vector(x,1,npar);
4187: free_ivector(indx,1,npar);
1.203 brouard 4188: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4189:
4190:
4191: }
4192:
4193: /*************** hessian matrix ****************/
4194: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4195: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4196: int i;
4197: int l=1, lmax=20;
1.203 brouard 4198: double k1,k2, res, fx;
1.132 brouard 4199: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4200: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4201: int k=0,kmax=10;
4202: double l1;
4203:
4204: fx=func(x);
4205: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4206: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4207: l1=pow(10,l);
4208: delts=delt;
4209: for(k=1 ; k <kmax; k=k+1){
4210: delt = delta*(l1*k);
4211: p2[theta]=x[theta] +delt;
1.145 brouard 4212: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4213: p2[theta]=x[theta]-delt;
4214: k2=func(p2)-fx;
4215: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4216: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4217:
1.203 brouard 4218: #ifdef DEBUGHESSII
1.126 brouard 4219: 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);
4220: 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);
4221: #endif
4222: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4223: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4224: k=kmax;
4225: }
4226: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4227: k=kmax; l=lmax*10;
1.126 brouard 4228: }
4229: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4230: delts=delt;
4231: }
1.203 brouard 4232: } /* End loop k */
1.126 brouard 4233: }
4234: delti[theta]=delts;
4235: return res;
4236:
4237: }
4238:
1.203 brouard 4239: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4240: {
4241: int i;
1.164 brouard 4242: int l=1, lmax=20;
1.126 brouard 4243: double k1,k2,k3,k4,res,fx;
1.132 brouard 4244: double p2[MAXPARM+1];
1.203 brouard 4245: int k, kmax=1;
4246: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4247:
4248: int firstime=0;
1.203 brouard 4249:
1.126 brouard 4250: fx=func(x);
1.203 brouard 4251: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4252: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4253: p2[thetai]=x[thetai]+delti[thetai]*k;
4254: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4255: k1=func(p2)-fx;
4256:
1.203 brouard 4257: p2[thetai]=x[thetai]+delti[thetai]*k;
4258: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4259: k2=func(p2)-fx;
4260:
1.203 brouard 4261: p2[thetai]=x[thetai]-delti[thetai]*k;
4262: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4263: k3=func(p2)-fx;
4264:
1.203 brouard 4265: p2[thetai]=x[thetai]-delti[thetai]*k;
4266: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4267: k4=func(p2)-fx;
1.203 brouard 4268: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4269: if(k1*k2*k3*k4 <0.){
1.208 brouard 4270: firstime=1;
1.203 brouard 4271: kmax=kmax+10;
1.208 brouard 4272: }
4273: if(kmax >=10 || firstime ==1){
1.246 brouard 4274: 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);
4275: 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 4276: 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);
4277: 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);
4278: }
4279: #ifdef DEBUGHESSIJ
4280: v1=hess[thetai][thetai];
4281: v2=hess[thetaj][thetaj];
4282: cv12=res;
4283: /* Computing eigen value of Hessian matrix */
4284: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4285: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4286: if ((lc2 <0) || (lc1 <0) ){
4287: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4288: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4289: 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);
4290: 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);
4291: }
1.126 brouard 4292: #endif
4293: }
4294: return res;
4295: }
4296:
1.203 brouard 4297: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4298: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4299: /* { */
4300: /* int i; */
4301: /* int l=1, lmax=20; */
4302: /* double k1,k2,k3,k4,res,fx; */
4303: /* double p2[MAXPARM+1]; */
4304: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4305: /* int k=0,kmax=10; */
4306: /* double l1; */
4307:
4308: /* fx=func(x); */
4309: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4310: /* l1=pow(10,l); */
4311: /* delts=delt; */
4312: /* for(k=1 ; k <kmax; k=k+1){ */
4313: /* delt = delti*(l1*k); */
4314: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4315: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4316: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4317: /* k1=func(p2)-fx; */
4318:
4319: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4320: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4321: /* k2=func(p2)-fx; */
4322:
4323: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4324: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4325: /* k3=func(p2)-fx; */
4326:
4327: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4328: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4329: /* k4=func(p2)-fx; */
4330: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4331: /* #ifdef DEBUGHESSIJ */
4332: /* 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); */
4333: /* 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); */
4334: /* #endif */
4335: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4336: /* k=kmax; */
4337: /* } */
4338: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4339: /* k=kmax; l=lmax*10; */
4340: /* } */
4341: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4342: /* delts=delt; */
4343: /* } */
4344: /* } /\* End loop k *\/ */
4345: /* } */
4346: /* delti[theta]=delts; */
4347: /* return res; */
4348: /* } */
4349:
4350:
1.126 brouard 4351: /************** Inverse of matrix **************/
4352: void ludcmp(double **a, int n, int *indx, double *d)
4353: {
4354: int i,imax,j,k;
4355: double big,dum,sum,temp;
4356: double *vv;
4357:
4358: vv=vector(1,n);
4359: *d=1.0;
4360: for (i=1;i<=n;i++) {
4361: big=0.0;
4362: for (j=1;j<=n;j++)
4363: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4364: if (big == 0.0){
4365: printf(" Singular Hessian matrix at row %d:\n",i);
4366: for (j=1;j<=n;j++) {
4367: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4368: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4369: }
4370: fflush(ficlog);
4371: fclose(ficlog);
4372: nrerror("Singular matrix in routine ludcmp");
4373: }
1.126 brouard 4374: vv[i]=1.0/big;
4375: }
4376: for (j=1;j<=n;j++) {
4377: for (i=1;i<j;i++) {
4378: sum=a[i][j];
4379: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4380: a[i][j]=sum;
4381: }
4382: big=0.0;
4383: for (i=j;i<=n;i++) {
4384: sum=a[i][j];
4385: for (k=1;k<j;k++)
4386: sum -= a[i][k]*a[k][j];
4387: a[i][j]=sum;
4388: if ( (dum=vv[i]*fabs(sum)) >= big) {
4389: big=dum;
4390: imax=i;
4391: }
4392: }
4393: if (j != imax) {
4394: for (k=1;k<=n;k++) {
4395: dum=a[imax][k];
4396: a[imax][k]=a[j][k];
4397: a[j][k]=dum;
4398: }
4399: *d = -(*d);
4400: vv[imax]=vv[j];
4401: }
4402: indx[j]=imax;
4403: if (a[j][j] == 0.0) a[j][j]=TINY;
4404: if (j != n) {
4405: dum=1.0/(a[j][j]);
4406: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4407: }
4408: }
4409: free_vector(vv,1,n); /* Doesn't work */
4410: ;
4411: }
4412:
4413: void lubksb(double **a, int n, int *indx, double b[])
4414: {
4415: int i,ii=0,ip,j;
4416: double sum;
4417:
4418: for (i=1;i<=n;i++) {
4419: ip=indx[i];
4420: sum=b[ip];
4421: b[ip]=b[i];
4422: if (ii)
4423: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4424: else if (sum) ii=i;
4425: b[i]=sum;
4426: }
4427: for (i=n;i>=1;i--) {
4428: sum=b[i];
4429: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4430: b[i]=sum/a[i][i];
4431: }
4432: }
4433:
4434: void pstamp(FILE *fichier)
4435: {
1.196 brouard 4436: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4437: }
4438:
1.297 brouard 4439: void date2dmy(double date,double *day, double *month, double *year){
4440: double yp=0., yp1=0., yp2=0.;
4441:
4442: yp1=modf(date,&yp);/* extracts integral of date in yp and
4443: fractional in yp1 */
4444: *year=yp;
4445: yp2=modf((yp1*12),&yp);
4446: *month=yp;
4447: yp1=modf((yp2*30.5),&yp);
4448: *day=yp;
4449: if(*day==0) *day=1;
4450: if(*month==0) *month=1;
4451: }
4452:
1.253 brouard 4453:
4454:
1.126 brouard 4455: /************ Frequencies ********************/
1.251 brouard 4456: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4457: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4458: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4459: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4460:
1.265 brouard 4461: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4462: int iind=0, iage=0;
4463: int mi; /* Effective wave */
4464: int first;
4465: double ***freq; /* Frequencies */
1.268 brouard 4466: 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 */
4467: 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);
1.284 brouard 4468: double *meanq, *stdq, *idq;
1.226 brouard 4469: double **meanqt;
4470: double *pp, **prop, *posprop, *pospropt;
4471: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4472: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4473: double agebegin, ageend;
4474:
4475: pp=vector(1,nlstate);
1.251 brouard 4476: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4477: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4478: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4479: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4480: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4481: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4482: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4483: meanqt=matrix(1,lastpass,1,nqtveff);
4484: strcpy(fileresp,"P_");
4485: strcat(fileresp,fileresu);
4486: /*strcat(fileresphtm,fileresu);*/
4487: if((ficresp=fopen(fileresp,"w"))==NULL) {
4488: printf("Problem with prevalence resultfile: %s\n", fileresp);
4489: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4490: exit(0);
4491: }
1.240 brouard 4492:
1.226 brouard 4493: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4494: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4495: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4496: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4497: fflush(ficlog);
4498: exit(70);
4499: }
4500: else{
4501: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4502: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4503: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4504: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4505: }
1.237 brouard 4506: 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 4507:
1.226 brouard 4508: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4509: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4510: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4511: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4512: fflush(ficlog);
4513: exit(70);
1.240 brouard 4514: } else{
1.226 brouard 4515: 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 4516: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4517: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4518: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4519: }
1.240 brouard 4520: 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);
4521:
1.253 brouard 4522: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4523: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4524: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4525: j1=0;
1.126 brouard 4526:
1.227 brouard 4527: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4528: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4529: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4530:
4531:
1.226 brouard 4532: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4533: reference=low_education V1=0,V2=0
4534: med_educ V1=1 V2=0,
4535: high_educ V1=0 V2=1
4536: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4537: */
1.249 brouard 4538: dateintsum=0;
4539: k2cpt=0;
4540:
1.253 brouard 4541: if(cptcoveff == 0 )
1.265 brouard 4542: nl=1; /* Constant and age model only */
1.253 brouard 4543: else
4544: nl=2;
1.265 brouard 4545:
4546: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4547: /* Loop on nj=1 or 2 if dummy covariates j!=0
4548: * Loop on j1(1 to 2**cptcoveff) covariate combination
4549: * freq[s1][s2][iage] =0.
4550: * Loop on iind
4551: * ++freq[s1][s2][iage] weighted
4552: * end iind
4553: * if covariate and j!0
4554: * headers Variable on one line
4555: * endif cov j!=0
4556: * header of frequency table by age
4557: * Loop on age
4558: * pp[s1]+=freq[s1][s2][iage] weighted
4559: * pos+=freq[s1][s2][iage] weighted
4560: * Loop on s1 initial state
4561: * fprintf(ficresp
4562: * end s1
4563: * end age
4564: * if j!=0 computes starting values
4565: * end compute starting values
4566: * end j1
4567: * end nl
4568: */
1.253 brouard 4569: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4570: if(nj==1)
4571: j=0; /* First pass for the constant */
1.265 brouard 4572: else{
1.253 brouard 4573: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4574: }
1.251 brouard 4575: first=1;
1.265 brouard 4576: 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 4577: posproptt=0.;
4578: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4579: scanf("%d", i);*/
4580: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4581: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4582: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4583: freq[i][s2][m]=0;
1.251 brouard 4584:
4585: for (i=1; i<=nlstate; i++) {
1.240 brouard 4586: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4587: prop[i][m]=0;
4588: posprop[i]=0;
4589: pospropt[i]=0;
4590: }
1.283 brouard 4591: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4592: idq[z1]=0.;
4593: meanq[z1]=0.;
4594: stdq[z1]=0.;
1.283 brouard 4595: }
4596: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4597: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4598: /* meanqt[m][z1]=0.; */
4599: /* } */
4600: /* } */
1.251 brouard 4601: /* dateintsum=0; */
4602: /* k2cpt=0; */
4603:
1.265 brouard 4604: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4605: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4606: bool=1;
4607: if(j !=0){
4608: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4609: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4610: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4611: /* if(Tvaraff[z1] ==-20){ */
4612: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4613: /* }else if(Tvaraff[z1] ==-10){ */
4614: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4615: /* }else */
4616: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4617: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4618: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4619: /* 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",
4620: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4621: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4622: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4623: } /* Onlyf fixed */
4624: } /* end z1 */
4625: } /* cptcovn > 0 */
4626: } /* end any */
4627: }/* end j==0 */
1.265 brouard 4628: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4629: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4630: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4631: m=mw[mi][iind];
4632: if(j!=0){
4633: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4634: for (z1=1; z1<=cptcoveff; z1++) {
4635: if( Fixed[Tmodelind[z1]]==1){
4636: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4637: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4638: value is -1, we don't select. It differs from the
4639: constant and age model which counts them. */
4640: bool=0; /* not selected */
4641: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4642: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4643: bool=0;
4644: }
4645: }
4646: }
4647: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4648: } /* end j==0 */
4649: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4650: if(bool==1){ /*Selected */
1.251 brouard 4651: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4652: and mw[mi+1][iind]. dh depends on stepm. */
4653: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4654: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4655: if(m >=firstpass && m <=lastpass){
4656: k2=anint[m][iind]+(mint[m][iind]/12.);
4657: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4658: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4659: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4660: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4661: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4662: if (m<lastpass) {
4663: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4664: /* 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]); */
4665: if(s[m][iind]==-1)
4666: 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.));
4667: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.284 brouard 4668: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4669: idq[z1]=idq[z1]+weight[iind];
4670: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4671: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4672: }
1.251 brouard 4673: /* if((int)agev[m][iind] == 55) */
4674: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4675: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4676: 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 4677: }
1.251 brouard 4678: } /* end if between passes */
4679: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4680: dateintsum=dateintsum+k2; /* on all covariates ?*/
4681: k2cpt++;
4682: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4683: }
1.251 brouard 4684: }else{
4685: bool=1;
4686: }/* end bool 2 */
4687: } /* end m */
1.284 brouard 4688: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4689: /* idq[z1]=idq[z1]+weight[iind]; */
4690: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4691: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4692: /* } */
1.251 brouard 4693: } /* end bool */
4694: } /* end iind = 1 to imx */
4695: /* prop[s][age] is feeded for any initial and valid live state as well as
4696: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4697:
4698:
4699: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4700: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4701: pstamp(ficresp);
1.251 brouard 4702: if (cptcoveff>0 && j!=0){
1.265 brouard 4703: pstamp(ficresp);
1.251 brouard 4704: printf( "\n#********** Variable ");
4705: fprintf(ficresp, "\n#********** Variable ");
4706: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4707: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4708: fprintf(ficlog, "\n#********** Variable ");
4709: for (z1=1; z1<=cptcoveff; z1++){
4710: if(!FixedV[Tvaraff[z1]]){
4711: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4712: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4713: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4714: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4715: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4716: }else{
1.251 brouard 4717: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4718: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4719: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4720: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4721: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4722: }
4723: }
4724: printf( "**********\n#");
4725: fprintf(ficresp, "**********\n#");
4726: fprintf(ficresphtm, "**********</h3>\n");
4727: fprintf(ficresphtmfr, "**********</h3>\n");
4728: fprintf(ficlog, "**********\n");
4729: }
1.284 brouard 4730: /*
4731: Printing means of quantitative variables if any
4732: */
4733: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4734: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4735: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4736: if(weightopt==1){
4737: printf(" Weighted mean and standard deviation of");
4738: fprintf(ficlog," Weighted mean and standard deviation of");
4739: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4740: }
1.285 brouard 4741: printf(" fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4742: fprintf(ficlog," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4743: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
1.284 brouard 4744: }
4745: /* for (z1=1; z1<= nqtveff; z1++) { */
4746: /* for(m=1;m<=lastpass;m++){ */
4747: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4748: /* } */
4749: /* } */
1.283 brouard 4750:
1.251 brouard 4751: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4752: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4753: fprintf(ficresp, " Age");
4754: 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 4755: for(i=1; i<=nlstate;i++) {
1.265 brouard 4756: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4757: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4758: }
1.265 brouard 4759: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4760: fprintf(ficresphtm, "\n");
4761:
4762: /* Header of frequency table by age */
4763: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4764: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4765: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4766: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4767: if(s2!=0 && m!=0)
4768: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4769: }
1.226 brouard 4770: }
1.251 brouard 4771: fprintf(ficresphtmfr, "\n");
4772:
4773: /* For each age */
4774: for(iage=iagemin; iage <= iagemax+3; iage++){
4775: fprintf(ficresphtm,"<tr>");
4776: if(iage==iagemax+1){
4777: fprintf(ficlog,"1");
4778: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4779: }else if(iage==iagemax+2){
4780: fprintf(ficlog,"0");
4781: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4782: }else if(iage==iagemax+3){
4783: fprintf(ficlog,"Total");
4784: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4785: }else{
1.240 brouard 4786: if(first==1){
1.251 brouard 4787: first=0;
4788: printf("See log file for details...\n");
4789: }
4790: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4791: fprintf(ficlog,"Age %d", iage);
4792: }
1.265 brouard 4793: for(s1=1; s1 <=nlstate ; s1++){
4794: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4795: pp[s1] += freq[s1][m][iage];
1.251 brouard 4796: }
1.265 brouard 4797: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4798: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4799: pos += freq[s1][m][iage];
4800: if(pp[s1]>=1.e-10){
1.251 brouard 4801: if(first==1){
1.265 brouard 4802: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4803: }
1.265 brouard 4804: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4805: }else{
4806: if(first==1)
1.265 brouard 4807: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4808: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4809: }
4810: }
4811:
1.265 brouard 4812: for(s1=1; s1 <=nlstate ; s1++){
4813: /* posprop[s1]=0; */
4814: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4815: pp[s1] += freq[s1][m][iage];
4816: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4817:
4818: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4819: pos += pp[s1]; /* pos is the total number of transitions until this age */
4820: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4821: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4822: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4823: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4824: }
4825:
4826: /* Writing ficresp */
4827: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4828: if( iage <= iagemax){
4829: fprintf(ficresp," %d",iage);
4830: }
4831: }else if( nj==2){
4832: if( iage <= iagemax){
4833: fprintf(ficresp," %d",iage);
4834: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4835: }
1.240 brouard 4836: }
1.265 brouard 4837: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4838: if(pos>=1.e-5){
1.251 brouard 4839: if(first==1)
1.265 brouard 4840: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4841: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4842: }else{
4843: if(first==1)
1.265 brouard 4844: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4845: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4846: }
4847: if( iage <= iagemax){
4848: if(pos>=1.e-5){
1.265 brouard 4849: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4850: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4851: }else if( nj==2){
4852: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4853: }
4854: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4855: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4856: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4857: } else{
4858: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4859: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4860: }
1.240 brouard 4861: }
1.265 brouard 4862: pospropt[s1] +=posprop[s1];
4863: } /* end loop s1 */
1.251 brouard 4864: /* pospropt=0.; */
1.265 brouard 4865: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4866: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4867: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4868: if(first==1){
1.265 brouard 4869: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4870: }
1.265 brouard 4871: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4872: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4873: }
1.265 brouard 4874: if(s1!=0 && m!=0)
4875: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4876: }
1.265 brouard 4877: } /* end loop s1 */
1.251 brouard 4878: posproptt=0.;
1.265 brouard 4879: for(s1=1; s1 <=nlstate; s1++){
4880: posproptt += pospropt[s1];
1.251 brouard 4881: }
4882: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4883: fprintf(ficresphtm,"</tr>\n");
4884: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4885: if(iage <= iagemax)
4886: fprintf(ficresp,"\n");
1.240 brouard 4887: }
1.251 brouard 4888: if(first==1)
4889: printf("Others in log...\n");
4890: fprintf(ficlog,"\n");
4891: } /* end loop age iage */
1.265 brouard 4892:
1.251 brouard 4893: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4894: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4895: if(posproptt < 1.e-5){
1.265 brouard 4896: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4897: }else{
1.265 brouard 4898: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4899: }
1.226 brouard 4900: }
1.251 brouard 4901: fprintf(ficresphtm,"</tr>\n");
4902: fprintf(ficresphtm,"</table>\n");
4903: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4904: if(posproptt < 1.e-5){
1.251 brouard 4905: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4906: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4907: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4908: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4909: invalidvarcomb[j1]=1;
1.226 brouard 4910: }else{
1.251 brouard 4911: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4912: invalidvarcomb[j1]=0;
1.226 brouard 4913: }
1.251 brouard 4914: fprintf(ficresphtmfr,"</table>\n");
4915: fprintf(ficlog,"\n");
4916: if(j!=0){
4917: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4918: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4919: for(k=1; k <=(nlstate+ndeath); k++){
4920: if (k != i) {
1.265 brouard 4921: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4922: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4923: if(j1==1){ /* All dummy covariates to zero */
4924: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4925: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4926: printf("%d%d ",i,k);
4927: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4928: 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]));
4929: 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]));
4930: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4931: }
1.253 brouard 4932: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4933: for(iage=iagemin; iage <= iagemax+3; iage++){
4934: x[iage]= (double)iage;
4935: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4936: /* 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 4937: }
1.268 brouard 4938: /* Some are not finite, but linreg will ignore these ages */
4939: no=0;
1.253 brouard 4940: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4941: pstart[s1]=b;
4942: pstart[s1-1]=a;
1.252 brouard 4943: }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 */
4944: 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]);
4945: 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 4946: 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 4947: printf("%d%d ",i,k);
4948: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4949: 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 4950: }else{ /* Other cases, like quantitative fixed or varying covariates */
4951: ;
4952: }
4953: /* printf("%12.7f )", param[i][jj][k]); */
4954: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4955: s1++;
1.251 brouard 4956: } /* end jj */
4957: } /* end k!= i */
4958: } /* end k */
1.265 brouard 4959: } /* end i, s1 */
1.251 brouard 4960: } /* end j !=0 */
4961: } /* end selected combination of covariate j1 */
4962: if(j==0){ /* We can estimate starting values from the occurences in each case */
4963: printf("#Freqsummary: Starting values for the constants:\n");
4964: fprintf(ficlog,"\n");
1.265 brouard 4965: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4966: for(k=1; k <=(nlstate+ndeath); k++){
4967: if (k != i) {
4968: printf("%d%d ",i,k);
4969: fprintf(ficlog,"%d%d ",i,k);
4970: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4971: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4972: if(jj==1){ /* Age has to be done */
1.265 brouard 4973: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4974: 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]));
4975: 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 4976: }
4977: /* printf("%12.7f )", param[i][jj][k]); */
4978: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4979: s1++;
1.250 brouard 4980: }
1.251 brouard 4981: printf("\n");
4982: fprintf(ficlog,"\n");
1.250 brouard 4983: }
4984: }
1.284 brouard 4985: } /* end of state i */
1.251 brouard 4986: printf("#Freqsummary\n");
4987: fprintf(ficlog,"\n");
1.265 brouard 4988: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4989: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4990: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4991: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4992: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4993: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4994: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4995: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4996: /* } */
4997: }
1.265 brouard 4998: } /* end loop s1 */
1.251 brouard 4999:
5000: printf("\n");
5001: fprintf(ficlog,"\n");
5002: } /* end j=0 */
1.249 brouard 5003: } /* end j */
1.252 brouard 5004:
1.253 brouard 5005: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5006: for(i=1, jk=1; i <=nlstate; i++){
5007: for(j=1; j <=nlstate+ndeath; j++){
5008: if(j!=i){
5009: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5010: printf("%1d%1d",i,j);
5011: fprintf(ficparo,"%1d%1d",i,j);
5012: for(k=1; k<=ncovmodel;k++){
5013: /* printf(" %lf",param[i][j][k]); */
5014: /* fprintf(ficparo," %lf",param[i][j][k]); */
5015: p[jk]=pstart[jk];
5016: printf(" %f ",pstart[jk]);
5017: fprintf(ficparo," %f ",pstart[jk]);
5018: jk++;
5019: }
5020: printf("\n");
5021: fprintf(ficparo,"\n");
5022: }
5023: }
5024: }
5025: } /* end mle=-2 */
1.226 brouard 5026: dateintmean=dateintsum/k2cpt;
1.296 brouard 5027: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5028:
1.226 brouard 5029: fclose(ficresp);
5030: fclose(ficresphtm);
5031: fclose(ficresphtmfr);
1.283 brouard 5032: free_vector(idq,1,nqfveff);
1.226 brouard 5033: free_vector(meanq,1,nqfveff);
1.284 brouard 5034: free_vector(stdq,1,nqfveff);
1.226 brouard 5035: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5036: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5037: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5038: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5039: free_vector(pospropt,1,nlstate);
5040: free_vector(posprop,1,nlstate);
1.251 brouard 5041: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5042: free_vector(pp,1,nlstate);
5043: /* End of freqsummary */
5044: }
1.126 brouard 5045:
1.268 brouard 5046: /* Simple linear regression */
5047: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5048:
5049: /* y=a+bx regression */
5050: double sumx = 0.0; /* sum of x */
5051: double sumx2 = 0.0; /* sum of x**2 */
5052: double sumxy = 0.0; /* sum of x * y */
5053: double sumy = 0.0; /* sum of y */
5054: double sumy2 = 0.0; /* sum of y**2 */
5055: double sume2 = 0.0; /* sum of square or residuals */
5056: double yhat;
5057:
5058: double denom=0;
5059: int i;
5060: int ne=*no;
5061:
5062: for ( i=ifi, ne=0;i<=ila;i++) {
5063: if(!isfinite(x[i]) || !isfinite(y[i])){
5064: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5065: continue;
5066: }
5067: ne=ne+1;
5068: sumx += x[i];
5069: sumx2 += x[i]*x[i];
5070: sumxy += x[i] * y[i];
5071: sumy += y[i];
5072: sumy2 += y[i]*y[i];
5073: denom = (ne * sumx2 - sumx*sumx);
5074: /* 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); */
5075: }
5076:
5077: denom = (ne * sumx2 - sumx*sumx);
5078: if (denom == 0) {
5079: // vertical, slope m is infinity
5080: *b = INFINITY;
5081: *a = 0;
5082: if (r) *r = 0;
5083: return 1;
5084: }
5085:
5086: *b = (ne * sumxy - sumx * sumy) / denom;
5087: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5088: if (r!=NULL) {
5089: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5090: sqrt((sumx2 - sumx*sumx/ne) *
5091: (sumy2 - sumy*sumy/ne));
5092: }
5093: *no=ne;
5094: for ( i=ifi, ne=0;i<=ila;i++) {
5095: if(!isfinite(x[i]) || !isfinite(y[i])){
5096: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5097: continue;
5098: }
5099: ne=ne+1;
5100: yhat = y[i] - *a -*b* x[i];
5101: sume2 += yhat * yhat ;
5102:
5103: denom = (ne * sumx2 - sumx*sumx);
5104: /* 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); */
5105: }
5106: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5107: *sa= *sb * sqrt(sumx2/ne);
5108:
5109: return 0;
5110: }
5111:
1.126 brouard 5112: /************ Prevalence ********************/
1.227 brouard 5113: 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)
5114: {
5115: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5116: in each health status at the date of interview (if between dateprev1 and dateprev2).
5117: We still use firstpass and lastpass as another selection.
5118: */
1.126 brouard 5119:
1.227 brouard 5120: int i, m, jk, j1, bool, z1,j, iv;
5121: int mi; /* Effective wave */
5122: int iage;
5123: double agebegin, ageend;
5124:
5125: double **prop;
5126: double posprop;
5127: double y2; /* in fractional years */
5128: int iagemin, iagemax;
5129: int first; /** to stop verbosity which is redirected to log file */
5130:
5131: iagemin= (int) agemin;
5132: iagemax= (int) agemax;
5133: /*pp=vector(1,nlstate);*/
1.251 brouard 5134: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5135: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5136: j1=0;
1.222 brouard 5137:
1.227 brouard 5138: /*j=cptcoveff;*/
5139: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5140:
1.288 brouard 5141: first=0;
1.227 brouard 5142: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5143: for (i=1; i<=nlstate; i++)
1.251 brouard 5144: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5145: prop[i][iage]=0.0;
5146: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5147: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5148: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5149:
5150: for (i=1; i<=imx; i++) { /* Each individual */
5151: bool=1;
5152: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5153: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5154: m=mw[mi][i];
5155: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5156: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5157: for (z1=1; z1<=cptcoveff; z1++){
5158: if( Fixed[Tmodelind[z1]]==1){
5159: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5160: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5161: bool=0;
5162: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5163: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5164: bool=0;
5165: }
5166: }
5167: if(bool==1){ /* Otherwise we skip that wave/person */
5168: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5169: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5170: if(m >=firstpass && m <=lastpass){
5171: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5172: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5173: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5174: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5175: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5176: 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);
5177: exit(1);
5178: }
5179: if (s[m][i]>0 && s[m][i]<=nlstate) {
5180: /*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]]);*/
5181: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5182: prop[s[m][i]][iagemax+3] += weight[i];
5183: } /* end valid statuses */
5184: } /* end selection of dates */
5185: } /* end selection of waves */
5186: } /* end bool */
5187: } /* end wave */
5188: } /* end individual */
5189: for(i=iagemin; i <= iagemax+3; i++){
5190: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5191: posprop += prop[jk][i];
5192: }
5193:
5194: for(jk=1; jk <=nlstate ; jk++){
5195: if( i <= iagemax){
5196: if(posprop>=1.e-5){
5197: probs[i][jk][j1]= prop[jk][i]/posprop;
5198: } else{
1.288 brouard 5199: if(!first){
5200: first=1;
1.266 brouard 5201: 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]);
5202: }else{
1.288 brouard 5203: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases.\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 5204: }
5205: }
5206: }
5207: }/* end jk */
5208: }/* end i */
1.222 brouard 5209: /*} *//* end i1 */
1.227 brouard 5210: } /* end j1 */
1.222 brouard 5211:
1.227 brouard 5212: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5213: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5214: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5215: } /* End of prevalence */
1.126 brouard 5216:
5217: /************* Waves Concatenation ***************/
5218:
5219: 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)
5220: {
1.298 brouard 5221: /* Concatenates waves: wav[i] is the number of effective (useful waves in the sense that a non interview is useless) of individual i.
1.126 brouard 5222: Death is a valid wave (if date is known).
5223: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5224: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5225: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5226: */
1.126 brouard 5227:
1.224 brouard 5228: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5229: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5230: double sum=0., jmean=0.;*/
1.224 brouard 5231: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5232: int j, k=0,jk, ju, jl;
5233: double sum=0.;
5234: first=0;
1.214 brouard 5235: firstwo=0;
1.217 brouard 5236: firsthree=0;
1.218 brouard 5237: firstfour=0;
1.164 brouard 5238: jmin=100000;
1.126 brouard 5239: jmax=-1;
5240: jmean=0.;
1.224 brouard 5241:
5242: /* Treating live states */
1.214 brouard 5243: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5244: mi=0; /* First valid wave */
1.227 brouard 5245: mli=0; /* Last valid wave */
1.126 brouard 5246: m=firstpass;
1.214 brouard 5247: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5248: 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 */
5249: mli=m-1;/* mw[++mi][i]=m-1; */
5250: }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 */
5251: mw[++mi][i]=m;
5252: mli=m;
1.224 brouard 5253: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5254: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5255: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5256: }
1.227 brouard 5257: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5258: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5259: break;
1.224 brouard 5260: #else
1.227 brouard 5261: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5262: if(firsthree == 0){
1.302 ! brouard 5263: 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 5264: firsthree=1;
5265: }
1.302 ! brouard 5266: 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 5267: mw[++mi][i]=m;
5268: mli=m;
5269: }
5270: if(s[m][i]==-2){ /* Vital status is really unknown */
5271: nbwarn++;
5272: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5273: 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);
5274: 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);
5275: }
5276: break;
5277: }
5278: break;
1.224 brouard 5279: #endif
1.227 brouard 5280: }/* End m >= lastpass */
1.126 brouard 5281: }/* end while */
1.224 brouard 5282:
1.227 brouard 5283: /* 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 5284: /* After last pass */
1.224 brouard 5285: /* Treating death states */
1.214 brouard 5286: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5287: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5288: /* } */
1.126 brouard 5289: mi++; /* Death is another wave */
5290: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5291: /* Only death is a correct wave */
1.126 brouard 5292: mw[mi][i]=m;
1.257 brouard 5293: } /* else not in a death state */
1.224 brouard 5294: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5295: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5296: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5297: 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 */
5298: nbwarn++;
5299: if(firstfiv==0){
5300: 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 );
5301: firstfiv=1;
5302: }else{
5303: 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 );
5304: }
5305: }else{ /* Death occured afer last wave potential bias */
5306: nberr++;
5307: if(firstwo==0){
1.257 brouard 5308: 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 5309: firstwo=1;
5310: }
1.257 brouard 5311: 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 5312: }
1.257 brouard 5313: }else{ /* if date of interview is unknown */
1.227 brouard 5314: /* death is known but not confirmed by death status at any wave */
5315: if(firstfour==0){
5316: 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 );
5317: firstfour=1;
5318: }
5319: 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 5320: }
1.224 brouard 5321: } /* end if date of death is known */
5322: #endif
5323: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5324: /* wav[i]=mw[mi][i]; */
1.126 brouard 5325: if(mi==0){
5326: nbwarn++;
5327: if(first==0){
1.227 brouard 5328: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5329: first=1;
1.126 brouard 5330: }
5331: if(first==1){
1.227 brouard 5332: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5333: }
5334: } /* end mi==0 */
5335: } /* End individuals */
1.214 brouard 5336: /* wav and mw are no more changed */
1.223 brouard 5337:
1.214 brouard 5338:
1.126 brouard 5339: for(i=1; i<=imx; i++){
5340: for(mi=1; mi<wav[i];mi++){
5341: if (stepm <=0)
1.227 brouard 5342: dh[mi][i]=1;
1.126 brouard 5343: else{
1.260 brouard 5344: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5345: if (agedc[i] < 2*AGESUP) {
5346: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5347: if(j==0) j=1; /* Survives at least one month after exam */
5348: else if(j<0){
5349: nberr++;
5350: 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]);
5351: j=1; /* Temporary Dangerous patch */
5352: 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);
5353: 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]);
5354: 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);
5355: }
5356: k=k+1;
5357: if (j >= jmax){
5358: jmax=j;
5359: ijmax=i;
5360: }
5361: if (j <= jmin){
5362: jmin=j;
5363: ijmin=i;
5364: }
5365: sum=sum+j;
5366: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5367: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5368: }
5369: }
5370: else{
5371: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5372: /* 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 5373:
1.227 brouard 5374: k=k+1;
5375: if (j >= jmax) {
5376: jmax=j;
5377: ijmax=i;
5378: }
5379: else if (j <= jmin){
5380: jmin=j;
5381: ijmin=i;
5382: }
5383: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5384: /*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]);*/
5385: if(j<0){
5386: nberr++;
5387: 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]);
5388: 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]);
5389: }
5390: sum=sum+j;
5391: }
5392: jk= j/stepm;
5393: jl= j -jk*stepm;
5394: ju= j -(jk+1)*stepm;
5395: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5396: if(jl==0){
5397: dh[mi][i]=jk;
5398: bh[mi][i]=0;
5399: }else{ /* We want a negative bias in order to only have interpolation ie
5400: * to avoid the price of an extra matrix product in likelihood */
5401: dh[mi][i]=jk+1;
5402: bh[mi][i]=ju;
5403: }
5404: }else{
5405: if(jl <= -ju){
5406: dh[mi][i]=jk;
5407: bh[mi][i]=jl; /* bias is positive if real duration
5408: * is higher than the multiple of stepm and negative otherwise.
5409: */
5410: }
5411: else{
5412: dh[mi][i]=jk+1;
5413: bh[mi][i]=ju;
5414: }
5415: if(dh[mi][i]==0){
5416: dh[mi][i]=1; /* At least one step */
5417: bh[mi][i]=ju; /* At least one step */
5418: /* 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);*/
5419: }
5420: } /* end if mle */
1.126 brouard 5421: }
5422: } /* end wave */
5423: }
5424: jmean=sum/k;
5425: 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 5426: 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 5427: }
1.126 brouard 5428:
5429: /*********** Tricode ****************************/
1.220 brouard 5430: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5431: {
5432: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5433: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5434: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5435: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5436: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5437: */
1.130 brouard 5438:
1.242 brouard 5439: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5440: int modmaxcovj=0; /* Modality max of covariates j */
5441: int cptcode=0; /* Modality max of covariates j */
5442: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5443:
5444:
1.242 brouard 5445: /* cptcoveff=0; */
5446: /* *cptcov=0; */
1.126 brouard 5447:
1.242 brouard 5448: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5449: for (k=1; k <= maxncov; k++)
5450: for(j=1; j<=2; j++)
5451: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5452:
1.242 brouard 5453: /* Loop on covariates without age and products and no quantitative variable */
5454: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5455: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5456: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5457: switch(Fixed[k]) {
5458: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5459: 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*/
5460: ij=(int)(covar[Tvar[k]][i]);
5461: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5462: * If product of Vn*Vm, still boolean *:
5463: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5464: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5465: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5466: modality of the nth covariate of individual i. */
5467: if (ij > modmaxcovj)
5468: modmaxcovj=ij;
5469: else if (ij < modmincovj)
5470: modmincovj=ij;
1.287 brouard 5471: if (ij <0 || ij >1 ){
5472: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5473: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5474: }
5475: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5476: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5477: exit(1);
5478: }else
5479: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5480: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5481: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5482: /* getting the maximum value of the modality of the covariate
5483: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5484: female ies 1, then modmaxcovj=1.
5485: */
5486: } /* end for loop on individuals i */
5487: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5488: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5489: cptcode=modmaxcovj;
5490: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5491: /*for (i=0; i<=cptcode; i++) {*/
5492: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5493: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5494: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5495: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5496: if( j != -1){
5497: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5498: covariate for which somebody answered excluding
5499: undefined. Usually 2: 0 and 1. */
5500: }
5501: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5502: covariate for which somebody answered including
5503: undefined. Usually 3: -1, 0 and 1. */
5504: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5505: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5506: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5507:
1.242 brouard 5508: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5509: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5510: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5511: /* modmincovj=3; modmaxcovj = 7; */
5512: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5513: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5514: /* defining two dummy variables: variables V1_1 and V1_2.*/
5515: /* nbcode[Tvar[j]][ij]=k; */
5516: /* nbcode[Tvar[j]][1]=0; */
5517: /* nbcode[Tvar[j]][2]=1; */
5518: /* nbcode[Tvar[j]][3]=2; */
5519: /* To be continued (not working yet). */
5520: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5521:
5522: /* 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*/
5523: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5524: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5525: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5526: /*, could be restored in the future */
5527: for (i=0; i<=1; i++) { /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
1.242 brouard 5528: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5529: break;
5530: }
5531: ij++;
1.287 brouard 5532: 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 . Could be -1*/
1.242 brouard 5533: cptcode = ij; /* New max modality for covar j */
5534: } /* end of loop on modality i=-1 to 1 or more */
5535: break;
5536: case 1: /* Testing on varying covariate, could be simple and
5537: * should look at waves or product of fixed *
5538: * varying. No time to test -1, assuming 0 and 1 only */
5539: ij=0;
5540: for(i=0; i<=1;i++){
5541: nbcode[Tvar[k]][++ij]=i;
5542: }
5543: break;
5544: default:
5545: break;
5546: } /* end switch */
5547: } /* end dummy test */
1.287 brouard 5548: } /* end of loop on model-covariate k. nbcode[Tvark][1]=-1, nbcode[Tvark][1]=0 and nbcode[Tvark][2]=1 sets the value of covariate k*/
1.242 brouard 5549:
5550: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5551: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5552: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5553: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5554: 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 */
5555: 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 */
5556: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5557: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5558:
5559: ij=0;
5560: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5561: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5562: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5563: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5564: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5565: /* If product not in single variable we don't print results */
5566: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5567: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5568: 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*/
5569: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5570: 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 */
5571: if(Fixed[k]!=0)
5572: anyvaryingduminmodel=1;
5573: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5574: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5575: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5576: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5577: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5578: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5579: }
5580: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5581: /* ij--; */
5582: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5583: *cptcov=ij; /*Number of total real effective covariates: effective
5584: * because they can be excluded from the model and real
5585: * if in the model but excluded because missing values, but how to get k from ij?*/
5586: for(j=ij+1; j<= cptcovt; j++){
5587: Tvaraff[j]=0;
5588: Tmodelind[j]=0;
5589: }
5590: for(j=ntveff+1; j<= cptcovt; j++){
5591: TmodelInvind[j]=0;
5592: }
5593: /* To be sorted */
5594: ;
5595: }
1.126 brouard 5596:
1.145 brouard 5597:
1.126 brouard 5598: /*********** Health Expectancies ****************/
5599:
1.235 brouard 5600: 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 5601:
5602: {
5603: /* Health expectancies, no variances */
1.164 brouard 5604: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5605: int nhstepma, nstepma; /* Decreasing with age */
5606: double age, agelim, hf;
5607: double ***p3mat;
5608: double eip;
5609:
1.238 brouard 5610: /* pstamp(ficreseij); */
1.126 brouard 5611: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5612: fprintf(ficreseij,"# Age");
5613: for(i=1; i<=nlstate;i++){
5614: for(j=1; j<=nlstate;j++){
5615: fprintf(ficreseij," e%1d%1d ",i,j);
5616: }
5617: fprintf(ficreseij," e%1d. ",i);
5618: }
5619: fprintf(ficreseij,"\n");
5620:
5621:
5622: if(estepm < stepm){
5623: printf ("Problem %d lower than %d\n",estepm, stepm);
5624: }
5625: else hstepm=estepm;
5626: /* We compute the life expectancy from trapezoids spaced every estepm months
5627: * This is mainly to measure the difference between two models: for example
5628: * if stepm=24 months pijx are given only every 2 years and by summing them
5629: * we are calculating an estimate of the Life Expectancy assuming a linear
5630: * progression in between and thus overestimating or underestimating according
5631: * to the curvature of the survival function. If, for the same date, we
5632: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5633: * to compare the new estimate of Life expectancy with the same linear
5634: * hypothesis. A more precise result, taking into account a more precise
5635: * curvature will be obtained if estepm is as small as stepm. */
5636:
5637: /* For example we decided to compute the life expectancy with the smallest unit */
5638: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5639: nhstepm is the number of hstepm from age to agelim
5640: nstepm is the number of stepm from age to agelin.
1.270 brouard 5641: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5642: and note for a fixed period like estepm months */
5643: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5644: survival function given by stepm (the optimization length). Unfortunately it
5645: means that if the survival funtion is printed only each two years of age and if
5646: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5647: results. So we changed our mind and took the option of the best precision.
5648: */
5649: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5650:
5651: agelim=AGESUP;
5652: /* If stepm=6 months */
5653: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5654: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5655:
5656: /* nhstepm age range expressed in number of stepm */
5657: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5658: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5659: /* if (stepm >= YEARM) hstepm=1;*/
5660: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5661: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5662:
5663: for (age=bage; age<=fage; age ++){
5664: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5665: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5666: /* if (stepm >= YEARM) hstepm=1;*/
5667: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5668:
5669: /* If stepm=6 months */
5670: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5671: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5672:
1.235 brouard 5673: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5674:
5675: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5676:
5677: printf("%d|",(int)age);fflush(stdout);
5678: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5679:
5680: /* Computing expectancies */
5681: for(i=1; i<=nlstate;i++)
5682: for(j=1; j<=nlstate;j++)
5683: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5684: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5685:
5686: /* 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]);*/
5687:
5688: }
5689:
5690: fprintf(ficreseij,"%3.0f",age );
5691: for(i=1; i<=nlstate;i++){
5692: eip=0;
5693: for(j=1; j<=nlstate;j++){
5694: eip +=eij[i][j][(int)age];
5695: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5696: }
5697: fprintf(ficreseij,"%9.4f", eip );
5698: }
5699: fprintf(ficreseij,"\n");
5700:
5701: }
5702: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5703: printf("\n");
5704: fprintf(ficlog,"\n");
5705:
5706: }
5707:
1.235 brouard 5708: 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 5709:
5710: {
5711: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5712: to initial status i, ei. .
1.126 brouard 5713: */
5714: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5715: int nhstepma, nstepma; /* Decreasing with age */
5716: double age, agelim, hf;
5717: double ***p3matp, ***p3matm, ***varhe;
5718: double **dnewm,**doldm;
5719: double *xp, *xm;
5720: double **gp, **gm;
5721: double ***gradg, ***trgradg;
5722: int theta;
5723:
5724: double eip, vip;
5725:
5726: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5727: xp=vector(1,npar);
5728: xm=vector(1,npar);
5729: dnewm=matrix(1,nlstate*nlstate,1,npar);
5730: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5731:
5732: pstamp(ficresstdeij);
5733: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5734: fprintf(ficresstdeij,"# Age");
5735: for(i=1; i<=nlstate;i++){
5736: for(j=1; j<=nlstate;j++)
5737: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5738: fprintf(ficresstdeij," e%1d. ",i);
5739: }
5740: fprintf(ficresstdeij,"\n");
5741:
5742: pstamp(ficrescveij);
5743: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5744: fprintf(ficrescveij,"# Age");
5745: for(i=1; i<=nlstate;i++)
5746: for(j=1; j<=nlstate;j++){
5747: cptj= (j-1)*nlstate+i;
5748: for(i2=1; i2<=nlstate;i2++)
5749: for(j2=1; j2<=nlstate;j2++){
5750: cptj2= (j2-1)*nlstate+i2;
5751: if(cptj2 <= cptj)
5752: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5753: }
5754: }
5755: fprintf(ficrescveij,"\n");
5756:
5757: if(estepm < stepm){
5758: printf ("Problem %d lower than %d\n",estepm, stepm);
5759: }
5760: else hstepm=estepm;
5761: /* We compute the life expectancy from trapezoids spaced every estepm months
5762: * This is mainly to measure the difference between two models: for example
5763: * if stepm=24 months pijx are given only every 2 years and by summing them
5764: * we are calculating an estimate of the Life Expectancy assuming a linear
5765: * progression in between and thus overestimating or underestimating according
5766: * to the curvature of the survival function. If, for the same date, we
5767: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5768: * to compare the new estimate of Life expectancy with the same linear
5769: * hypothesis. A more precise result, taking into account a more precise
5770: * curvature will be obtained if estepm is as small as stepm. */
5771:
5772: /* For example we decided to compute the life expectancy with the smallest unit */
5773: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5774: nhstepm is the number of hstepm from age to agelim
5775: nstepm is the number of stepm from age to agelin.
5776: Look at hpijx to understand the reason of that which relies in memory size
5777: and note for a fixed period like estepm months */
5778: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5779: survival function given by stepm (the optimization length). Unfortunately it
5780: means that if the survival funtion is printed only each two years of age and if
5781: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5782: results. So we changed our mind and took the option of the best precision.
5783: */
5784: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5785:
5786: /* If stepm=6 months */
5787: /* nhstepm age range expressed in number of stepm */
5788: agelim=AGESUP;
5789: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5790: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5791: /* if (stepm >= YEARM) hstepm=1;*/
5792: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5793:
5794: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5795: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5796: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5797: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5798: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5799: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5800:
5801: for (age=bage; age<=fage; age ++){
5802: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5803: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5804: /* if (stepm >= YEARM) hstepm=1;*/
5805: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5806:
1.126 brouard 5807: /* If stepm=6 months */
5808: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5809: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5810:
5811: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5812:
1.126 brouard 5813: /* Computing Variances of health expectancies */
5814: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5815: decrease memory allocation */
5816: for(theta=1; theta <=npar; theta++){
5817: for(i=1; i<=npar; i++){
1.222 brouard 5818: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5819: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5820: }
1.235 brouard 5821: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5822: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5823:
1.126 brouard 5824: for(j=1; j<= nlstate; j++){
1.222 brouard 5825: for(i=1; i<=nlstate; i++){
5826: for(h=0; h<=nhstepm-1; h++){
5827: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5828: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5829: }
5830: }
1.126 brouard 5831: }
1.218 brouard 5832:
1.126 brouard 5833: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5834: for(h=0; h<=nhstepm-1; h++){
5835: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5836: }
1.126 brouard 5837: }/* End theta */
5838:
5839:
5840: for(h=0; h<=nhstepm-1; h++)
5841: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5842: for(theta=1; theta <=npar; theta++)
5843: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5844:
1.218 brouard 5845:
1.222 brouard 5846: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5847: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5848: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5849:
1.222 brouard 5850: printf("%d|",(int)age);fflush(stdout);
5851: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5852: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5853: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5854: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5855: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5856: for(ij=1;ij<=nlstate*nlstate;ij++)
5857: for(ji=1;ji<=nlstate*nlstate;ji++)
5858: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5859: }
5860: }
1.218 brouard 5861:
1.126 brouard 5862: /* Computing expectancies */
1.235 brouard 5863: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5864: for(i=1; i<=nlstate;i++)
5865: for(j=1; j<=nlstate;j++)
1.222 brouard 5866: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5867: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5868:
1.222 brouard 5869: /* 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 5870:
1.222 brouard 5871: }
1.269 brouard 5872:
5873: /* Standard deviation of expectancies ij */
1.126 brouard 5874: fprintf(ficresstdeij,"%3.0f",age );
5875: for(i=1; i<=nlstate;i++){
5876: eip=0.;
5877: vip=0.;
5878: for(j=1; j<=nlstate;j++){
1.222 brouard 5879: eip += eij[i][j][(int)age];
5880: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5881: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5882: 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 5883: }
5884: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5885: }
5886: fprintf(ficresstdeij,"\n");
1.218 brouard 5887:
1.269 brouard 5888: /* Variance of expectancies ij */
1.126 brouard 5889: fprintf(ficrescveij,"%3.0f",age );
5890: for(i=1; i<=nlstate;i++)
5891: for(j=1; j<=nlstate;j++){
1.222 brouard 5892: cptj= (j-1)*nlstate+i;
5893: for(i2=1; i2<=nlstate;i2++)
5894: for(j2=1; j2<=nlstate;j2++){
5895: cptj2= (j2-1)*nlstate+i2;
5896: if(cptj2 <= cptj)
5897: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5898: }
1.126 brouard 5899: }
5900: fprintf(ficrescveij,"\n");
1.218 brouard 5901:
1.126 brouard 5902: }
5903: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5904: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5905: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5906: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5907: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5908: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5909: printf("\n");
5910: fprintf(ficlog,"\n");
1.218 brouard 5911:
1.126 brouard 5912: free_vector(xm,1,npar);
5913: free_vector(xp,1,npar);
5914: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5915: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5916: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5917: }
1.218 brouard 5918:
1.126 brouard 5919: /************ Variance ******************/
1.235 brouard 5920: 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 5921: {
1.279 brouard 5922: /** Variance of health expectancies
5923: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5924: * double **newm;
5925: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5926: */
1.218 brouard 5927:
5928: /* int movingaverage(); */
5929: double **dnewm,**doldm;
5930: double **dnewmp,**doldmp;
5931: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5932: int first=0;
1.218 brouard 5933: int k;
5934: double *xp;
1.279 brouard 5935: double **gp, **gm; /**< for var eij */
5936: double ***gradg, ***trgradg; /**< for var eij */
5937: double **gradgp, **trgradgp; /**< for var p point j */
5938: double *gpp, *gmp; /**< for var p point j */
5939: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5940: double ***p3mat;
5941: double age,agelim, hf;
5942: /* double ***mobaverage; */
5943: int theta;
5944: char digit[4];
5945: char digitp[25];
5946:
5947: char fileresprobmorprev[FILENAMELENGTH];
5948:
5949: if(popbased==1){
5950: if(mobilav!=0)
5951: strcpy(digitp,"-POPULBASED-MOBILAV_");
5952: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5953: }
5954: else
5955: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5956:
1.218 brouard 5957: /* if (mobilav!=0) { */
5958: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5959: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5960: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5961: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5962: /* } */
5963: /* } */
5964:
5965: strcpy(fileresprobmorprev,"PRMORPREV-");
5966: sprintf(digit,"%-d",ij);
5967: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5968: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5969: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5970: strcat(fileresprobmorprev,fileresu);
5971: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5972: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5973: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5974: }
5975: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5976: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5977: pstamp(ficresprobmorprev);
5978: 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 5979: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5980: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5981: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5982: }
5983: for(j=1;j<=cptcoveff;j++)
5984: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5985: fprintf(ficresprobmorprev,"\n");
5986:
1.218 brouard 5987: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5988: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5989: fprintf(ficresprobmorprev," p.%-d SE",j);
5990: for(i=1; i<=nlstate;i++)
5991: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5992: }
5993: fprintf(ficresprobmorprev,"\n");
5994:
5995: fprintf(ficgp,"\n# Routine varevsij");
5996: fprintf(ficgp,"\nunset title \n");
5997: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5998: 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");
5999: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6000:
1.218 brouard 6001: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6002: pstamp(ficresvij);
6003: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6004: if(popbased==1)
6005: 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);
6006: else
6007: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6008: fprintf(ficresvij,"# Age");
6009: for(i=1; i<=nlstate;i++)
6010: for(j=1; j<=nlstate;j++)
6011: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6012: fprintf(ficresvij,"\n");
6013:
6014: xp=vector(1,npar);
6015: dnewm=matrix(1,nlstate,1,npar);
6016: doldm=matrix(1,nlstate,1,nlstate);
6017: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6018: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6019:
6020: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6021: gpp=vector(nlstate+1,nlstate+ndeath);
6022: gmp=vector(nlstate+1,nlstate+ndeath);
6023: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6024:
1.218 brouard 6025: if(estepm < stepm){
6026: printf ("Problem %d lower than %d\n",estepm, stepm);
6027: }
6028: else hstepm=estepm;
6029: /* For example we decided to compute the life expectancy with the smallest unit */
6030: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6031: nhstepm is the number of hstepm from age to agelim
6032: nstepm is the number of stepm from age to agelim.
6033: Look at function hpijx to understand why because of memory size limitations,
6034: we decided (b) to get a life expectancy respecting the most precise curvature of the
6035: survival function given by stepm (the optimization length). Unfortunately it
6036: means that if the survival funtion is printed every two years of age and if
6037: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6038: results. So we changed our mind and took the option of the best precision.
6039: */
6040: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6041: agelim = AGESUP;
6042: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6043: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6044: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6045: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6046: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6047: gp=matrix(0,nhstepm,1,nlstate);
6048: gm=matrix(0,nhstepm,1,nlstate);
6049:
6050:
6051: for(theta=1; theta <=npar; theta++){
6052: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6053: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6054: }
1.279 brouard 6055: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6056: * returns into prlim .
1.288 brouard 6057: */
1.242 brouard 6058: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6059:
6060: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6061: if (popbased==1) {
6062: if(mobilav ==0){
6063: for(i=1; i<=nlstate;i++)
6064: prlim[i][i]=probs[(int)age][i][ij];
6065: }else{ /* mobilav */
6066: for(i=1; i<=nlstate;i++)
6067: prlim[i][i]=mobaverage[(int)age][i][ij];
6068: }
6069: }
1.295 brouard 6070: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6071: */
6072: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
1.292 brouard 6073: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}x\f$, which are the probability
1.279 brouard 6074: * at horizon h in state j including mortality.
6075: */
1.218 brouard 6076: for(j=1; j<= nlstate; j++){
6077: for(h=0; h<=nhstepm; h++){
6078: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6079: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6080: }
6081: }
1.279 brouard 6082: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6083: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6084: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6085: */
6086: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6087: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6088: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6089: }
6090:
6091: /* Again with minus shift */
1.218 brouard 6092:
6093: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6094: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6095:
1.242 brouard 6096: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6097:
6098: if (popbased==1) {
6099: if(mobilav ==0){
6100: for(i=1; i<=nlstate;i++)
6101: prlim[i][i]=probs[(int)age][i][ij];
6102: }else{ /* mobilav */
6103: for(i=1; i<=nlstate;i++)
6104: prlim[i][i]=mobaverage[(int)age][i][ij];
6105: }
6106: }
6107:
1.235 brouard 6108: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6109:
6110: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6111: for(h=0; h<=nhstepm; h++){
6112: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6113: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6114: }
6115: }
6116: /* This for computing probability of death (h=1 means
6117: computed over hstepm matrices product = hstepm*stepm months)
6118: as a weighted average of prlim.
6119: */
6120: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6121: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6122: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6123: }
1.279 brouard 6124: /* end shifting computations */
6125:
6126: /**< Computing gradient matrix at horizon h
6127: */
1.218 brouard 6128: for(j=1; j<= nlstate; j++) /* vareij */
6129: for(h=0; h<=nhstepm; h++){
6130: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6131: }
1.279 brouard 6132: /**< Gradient of overall mortality p.3 (or p.j)
6133: */
6134: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6135: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6136: }
6137:
6138: } /* End theta */
1.279 brouard 6139:
6140: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6141: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6142:
6143: for(h=0; h<=nhstepm; h++) /* veij */
6144: for(j=1; j<=nlstate;j++)
6145: for(theta=1; theta <=npar; theta++)
6146: trgradg[h][j][theta]=gradg[h][theta][j];
6147:
6148: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6149: for(theta=1; theta <=npar; theta++)
6150: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6151: /**< as well as its transposed matrix
6152: */
1.218 brouard 6153:
6154: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6155: for(i=1;i<=nlstate;i++)
6156: for(j=1;j<=nlstate;j++)
6157: vareij[i][j][(int)age] =0.;
1.279 brouard 6158:
6159: /* Computing trgradg by matcov by gradg at age and summing over h
6160: * and k (nhstepm) formula 15 of article
6161: * Lievre-Brouard-Heathcote
6162: */
6163:
1.218 brouard 6164: for(h=0;h<=nhstepm;h++){
6165: for(k=0;k<=nhstepm;k++){
6166: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6167: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6168: for(i=1;i<=nlstate;i++)
6169: for(j=1;j<=nlstate;j++)
6170: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6171: }
6172: }
6173:
1.279 brouard 6174: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6175: * p.j overall mortality formula 49 but computed directly because
6176: * we compute the grad (wix pijx) instead of grad (pijx),even if
6177: * wix is independent of theta.
6178: */
1.218 brouard 6179: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6180: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6181: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6182: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6183: varppt[j][i]=doldmp[j][i];
6184: /* end ppptj */
6185: /* x centered again */
6186:
1.242 brouard 6187: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6188:
6189: if (popbased==1) {
6190: if(mobilav ==0){
6191: for(i=1; i<=nlstate;i++)
6192: prlim[i][i]=probs[(int)age][i][ij];
6193: }else{ /* mobilav */
6194: for(i=1; i<=nlstate;i++)
6195: prlim[i][i]=mobaverage[(int)age][i][ij];
6196: }
6197: }
6198:
6199: /* This for computing probability of death (h=1 means
6200: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6201: as a weighted average of prlim.
6202: */
1.235 brouard 6203: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6204: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6205: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6206: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6207: }
6208: /* end probability of death */
6209:
6210: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6211: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6212: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6213: for(i=1; i<=nlstate;i++){
6214: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6215: }
6216: }
6217: fprintf(ficresprobmorprev,"\n");
6218:
6219: fprintf(ficresvij,"%.0f ",age );
6220: for(i=1; i<=nlstate;i++)
6221: for(j=1; j<=nlstate;j++){
6222: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6223: }
6224: fprintf(ficresvij,"\n");
6225: free_matrix(gp,0,nhstepm,1,nlstate);
6226: free_matrix(gm,0,nhstepm,1,nlstate);
6227: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6228: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6229: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6230: } /* End age */
6231: free_vector(gpp,nlstate+1,nlstate+ndeath);
6232: free_vector(gmp,nlstate+1,nlstate+ndeath);
6233: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6234: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6235: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6236: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6237: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6238: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6239: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6240: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6241: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6242: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6243: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6244: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6245: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6246: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6247: 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);
6248: /* 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 6249: */
1.218 brouard 6250: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6251: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6252:
1.218 brouard 6253: free_vector(xp,1,npar);
6254: free_matrix(doldm,1,nlstate,1,nlstate);
6255: free_matrix(dnewm,1,nlstate,1,npar);
6256: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6257: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6258: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6259: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6260: fclose(ficresprobmorprev);
6261: fflush(ficgp);
6262: fflush(fichtm);
6263: } /* end varevsij */
1.126 brouard 6264:
6265: /************ Variance of prevlim ******************/
1.269 brouard 6266: 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 6267: {
1.205 brouard 6268: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6269: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6270:
1.268 brouard 6271: double **dnewmpar,**doldm;
1.126 brouard 6272: int i, j, nhstepm, hstepm;
6273: double *xp;
6274: double *gp, *gm;
6275: double **gradg, **trgradg;
1.208 brouard 6276: double **mgm, **mgp;
1.126 brouard 6277: double age,agelim;
6278: int theta;
6279:
6280: pstamp(ficresvpl);
1.288 brouard 6281: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6282: fprintf(ficresvpl,"# Age ");
6283: if(nresult >=1)
6284: fprintf(ficresvpl," Result# ");
1.126 brouard 6285: for(i=1; i<=nlstate;i++)
6286: fprintf(ficresvpl," %1d-%1d",i,i);
6287: fprintf(ficresvpl,"\n");
6288:
6289: xp=vector(1,npar);
1.268 brouard 6290: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6291: doldm=matrix(1,nlstate,1,nlstate);
6292:
6293: hstepm=1*YEARM; /* Every year of age */
6294: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6295: agelim = AGESUP;
6296: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6297: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6298: if (stepm >= YEARM) hstepm=1;
6299: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6300: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6301: mgp=matrix(1,npar,1,nlstate);
6302: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6303: gp=vector(1,nlstate);
6304: gm=vector(1,nlstate);
6305:
6306: for(theta=1; theta <=npar; theta++){
6307: for(i=1; i<=npar; i++){ /* Computes gradient */
6308: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6309: }
1.288 brouard 6310: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6311: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6312: /* else */
6313: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6314: for(i=1;i<=nlstate;i++){
1.126 brouard 6315: gp[i] = prlim[i][i];
1.208 brouard 6316: mgp[theta][i] = prlim[i][i];
6317: }
1.126 brouard 6318: for(i=1; i<=npar; i++) /* Computes gradient */
6319: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6320: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6321: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6322: /* else */
6323: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6324: for(i=1;i<=nlstate;i++){
1.126 brouard 6325: gm[i] = prlim[i][i];
1.208 brouard 6326: mgm[theta][i] = prlim[i][i];
6327: }
1.126 brouard 6328: for(i=1;i<=nlstate;i++)
6329: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6330: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6331: } /* End theta */
6332:
6333: trgradg =matrix(1,nlstate,1,npar);
6334:
6335: for(j=1; j<=nlstate;j++)
6336: for(theta=1; theta <=npar; theta++)
6337: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6338: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6339: /* printf("\nmgm mgp %d ",(int)age); */
6340: /* for(j=1; j<=nlstate;j++){ */
6341: /* printf(" %d ",j); */
6342: /* for(theta=1; theta <=npar; theta++) */
6343: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6344: /* printf("\n "); */
6345: /* } */
6346: /* } */
6347: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6348: /* printf("\n gradg %d ",(int)age); */
6349: /* for(j=1; j<=nlstate;j++){ */
6350: /* printf("%d ",j); */
6351: /* for(theta=1; theta <=npar; theta++) */
6352: /* printf("%d %lf ",theta,gradg[theta][j]); */
6353: /* printf("\n "); */
6354: /* } */
6355: /* } */
1.126 brouard 6356:
6357: for(i=1;i<=nlstate;i++)
6358: varpl[i][(int)age] =0.;
1.209 brouard 6359: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6360: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6361: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6362: }else{
1.268 brouard 6363: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6364: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6365: }
1.126 brouard 6366: for(i=1;i<=nlstate;i++)
6367: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6368:
6369: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6370: if(nresult >=1)
6371: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6372: for(i=1; i<=nlstate;i++){
1.126 brouard 6373: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6374: /* for(j=1;j<=nlstate;j++) */
6375: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6376: }
1.126 brouard 6377: fprintf(ficresvpl,"\n");
6378: free_vector(gp,1,nlstate);
6379: free_vector(gm,1,nlstate);
1.208 brouard 6380: free_matrix(mgm,1,npar,1,nlstate);
6381: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6382: free_matrix(gradg,1,npar,1,nlstate);
6383: free_matrix(trgradg,1,nlstate,1,npar);
6384: } /* End age */
6385:
6386: free_vector(xp,1,npar);
6387: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6388: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6389:
6390: }
6391:
6392:
6393: /************ Variance of backprevalence limit ******************/
1.269 brouard 6394: 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 6395: {
6396: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6397: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6398:
6399: double **dnewmpar,**doldm;
6400: int i, j, nhstepm, hstepm;
6401: double *xp;
6402: double *gp, *gm;
6403: double **gradg, **trgradg;
6404: double **mgm, **mgp;
6405: double age,agelim;
6406: int theta;
6407:
6408: pstamp(ficresvbl);
6409: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6410: fprintf(ficresvbl,"# Age ");
6411: if(nresult >=1)
6412: fprintf(ficresvbl," Result# ");
6413: for(i=1; i<=nlstate;i++)
6414: fprintf(ficresvbl," %1d-%1d",i,i);
6415: fprintf(ficresvbl,"\n");
6416:
6417: xp=vector(1,npar);
6418: dnewmpar=matrix(1,nlstate,1,npar);
6419: doldm=matrix(1,nlstate,1,nlstate);
6420:
6421: hstepm=1*YEARM; /* Every year of age */
6422: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6423: agelim = AGEINF;
6424: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6425: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6426: if (stepm >= YEARM) hstepm=1;
6427: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6428: gradg=matrix(1,npar,1,nlstate);
6429: mgp=matrix(1,npar,1,nlstate);
6430: mgm=matrix(1,npar,1,nlstate);
6431: gp=vector(1,nlstate);
6432: gm=vector(1,nlstate);
6433:
6434: for(theta=1; theta <=npar; theta++){
6435: for(i=1; i<=npar; i++){ /* Computes gradient */
6436: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6437: }
6438: if(mobilavproj > 0 )
6439: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6440: else
6441: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6442: for(i=1;i<=nlstate;i++){
6443: gp[i] = bprlim[i][i];
6444: mgp[theta][i] = bprlim[i][i];
6445: }
6446: for(i=1; i<=npar; i++) /* Computes gradient */
6447: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6448: if(mobilavproj > 0 )
6449: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6450: else
6451: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6452: for(i=1;i<=nlstate;i++){
6453: gm[i] = bprlim[i][i];
6454: mgm[theta][i] = bprlim[i][i];
6455: }
6456: for(i=1;i<=nlstate;i++)
6457: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6458: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6459: } /* End theta */
6460:
6461: trgradg =matrix(1,nlstate,1,npar);
6462:
6463: for(j=1; j<=nlstate;j++)
6464: for(theta=1; theta <=npar; theta++)
6465: trgradg[j][theta]=gradg[theta][j];
6466: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6467: /* printf("\nmgm mgp %d ",(int)age); */
6468: /* for(j=1; j<=nlstate;j++){ */
6469: /* printf(" %d ",j); */
6470: /* for(theta=1; theta <=npar; theta++) */
6471: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6472: /* printf("\n "); */
6473: /* } */
6474: /* } */
6475: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6476: /* printf("\n gradg %d ",(int)age); */
6477: /* for(j=1; j<=nlstate;j++){ */
6478: /* printf("%d ",j); */
6479: /* for(theta=1; theta <=npar; theta++) */
6480: /* printf("%d %lf ",theta,gradg[theta][j]); */
6481: /* printf("\n "); */
6482: /* } */
6483: /* } */
6484:
6485: for(i=1;i<=nlstate;i++)
6486: varbpl[i][(int)age] =0.;
6487: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6488: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6489: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6490: }else{
6491: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6492: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6493: }
6494: for(i=1;i<=nlstate;i++)
6495: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6496:
6497: fprintf(ficresvbl,"%.0f ",age );
6498: if(nresult >=1)
6499: fprintf(ficresvbl,"%d ",nres );
6500: for(i=1; i<=nlstate;i++)
6501: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6502: fprintf(ficresvbl,"\n");
6503: free_vector(gp,1,nlstate);
6504: free_vector(gm,1,nlstate);
6505: free_matrix(mgm,1,npar,1,nlstate);
6506: free_matrix(mgp,1,npar,1,nlstate);
6507: free_matrix(gradg,1,npar,1,nlstate);
6508: free_matrix(trgradg,1,nlstate,1,npar);
6509: } /* End age */
6510:
6511: free_vector(xp,1,npar);
6512: free_matrix(doldm,1,nlstate,1,npar);
6513: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6514:
6515: }
6516:
6517: /************ Variance of one-step probabilities ******************/
6518: 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 6519: {
6520: int i, j=0, k1, l1, tj;
6521: int k2, l2, j1, z1;
6522: int k=0, l;
6523: int first=1, first1, first2;
6524: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6525: double **dnewm,**doldm;
6526: double *xp;
6527: double *gp, *gm;
6528: double **gradg, **trgradg;
6529: double **mu;
6530: double age, cov[NCOVMAX+1];
6531: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6532: int theta;
6533: char fileresprob[FILENAMELENGTH];
6534: char fileresprobcov[FILENAMELENGTH];
6535: char fileresprobcor[FILENAMELENGTH];
6536: double ***varpij;
6537:
6538: strcpy(fileresprob,"PROB_");
6539: strcat(fileresprob,fileres);
6540: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6541: printf("Problem with resultfile: %s\n", fileresprob);
6542: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6543: }
6544: strcpy(fileresprobcov,"PROBCOV_");
6545: strcat(fileresprobcov,fileresu);
6546: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6547: printf("Problem with resultfile: %s\n", fileresprobcov);
6548: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6549: }
6550: strcpy(fileresprobcor,"PROBCOR_");
6551: strcat(fileresprobcor,fileresu);
6552: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6553: printf("Problem with resultfile: %s\n", fileresprobcor);
6554: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6555: }
6556: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6557: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6558: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6559: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6560: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6561: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6562: pstamp(ficresprob);
6563: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6564: fprintf(ficresprob,"# Age");
6565: pstamp(ficresprobcov);
6566: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6567: fprintf(ficresprobcov,"# Age");
6568: pstamp(ficresprobcor);
6569: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6570: fprintf(ficresprobcor,"# Age");
1.126 brouard 6571:
6572:
1.222 brouard 6573: for(i=1; i<=nlstate;i++)
6574: for(j=1; j<=(nlstate+ndeath);j++){
6575: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6576: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6577: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6578: }
6579: /* fprintf(ficresprob,"\n");
6580: fprintf(ficresprobcov,"\n");
6581: fprintf(ficresprobcor,"\n");
6582: */
6583: xp=vector(1,npar);
6584: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6585: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6586: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6587: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6588: first=1;
6589: fprintf(ficgp,"\n# Routine varprob");
6590: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6591: fprintf(fichtm,"\n");
6592:
1.288 brouard 6593: 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. File %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6594: 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);
6595: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6596: and drawn. It helps understanding how is the covariance between two incidences.\
6597: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6598: 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 6599: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6600: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6601: standard deviations wide on each axis. <br>\
6602: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6603: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6604: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6605:
1.222 brouard 6606: cov[1]=1;
6607: /* tj=cptcoveff; */
1.225 brouard 6608: tj = (int) pow(2,cptcoveff);
1.222 brouard 6609: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6610: j1=0;
1.224 brouard 6611: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6612: if (cptcovn>0) {
6613: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6614: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6615: fprintf(ficresprob, "**********\n#\n");
6616: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6617: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6618: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6619:
1.222 brouard 6620: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6621: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6622: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6623:
6624:
1.222 brouard 6625: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6626: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6627: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6628:
1.222 brouard 6629: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6630: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6631: fprintf(ficresprobcor, "**********\n#");
6632: if(invalidvarcomb[j1]){
6633: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6634: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6635: continue;
6636: }
6637: }
6638: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6639: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6640: gp=vector(1,(nlstate)*(nlstate+ndeath));
6641: gm=vector(1,(nlstate)*(nlstate+ndeath));
6642: for (age=bage; age<=fage; age ++){
6643: cov[2]=age;
6644: if(nagesqr==1)
6645: cov[3]= age*age;
6646: for (k=1; k<=cptcovn;k++) {
6647: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6648: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6649: * 1 1 1 1 1
6650: * 2 2 1 1 1
6651: * 3 1 2 1 1
6652: */
6653: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6654: }
6655: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6656: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6657: for (k=1; k<=cptcovprod;k++)
6658: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6659:
6660:
1.222 brouard 6661: for(theta=1; theta <=npar; theta++){
6662: for(i=1; i<=npar; i++)
6663: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6664:
1.222 brouard 6665: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6666:
1.222 brouard 6667: k=0;
6668: for(i=1; i<= (nlstate); i++){
6669: for(j=1; j<=(nlstate+ndeath);j++){
6670: k=k+1;
6671: gp[k]=pmmij[i][j];
6672: }
6673: }
1.220 brouard 6674:
1.222 brouard 6675: for(i=1; i<=npar; i++)
6676: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6677:
1.222 brouard 6678: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6679: k=0;
6680: for(i=1; i<=(nlstate); i++){
6681: for(j=1; j<=(nlstate+ndeath);j++){
6682: k=k+1;
6683: gm[k]=pmmij[i][j];
6684: }
6685: }
1.220 brouard 6686:
1.222 brouard 6687: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6688: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6689: }
1.126 brouard 6690:
1.222 brouard 6691: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6692: for(theta=1; theta <=npar; theta++)
6693: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6694:
1.222 brouard 6695: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6696: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6697:
1.222 brouard 6698: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6699:
1.222 brouard 6700: k=0;
6701: for(i=1; i<=(nlstate); i++){
6702: for(j=1; j<=(nlstate+ndeath);j++){
6703: k=k+1;
6704: mu[k][(int) age]=pmmij[i][j];
6705: }
6706: }
6707: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6708: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6709: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6710:
1.222 brouard 6711: /*printf("\n%d ",(int)age);
6712: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6713: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6714: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6715: }*/
1.220 brouard 6716:
1.222 brouard 6717: fprintf(ficresprob,"\n%d ",(int)age);
6718: fprintf(ficresprobcov,"\n%d ",(int)age);
6719: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6720:
1.222 brouard 6721: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6722: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6723: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6724: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6725: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6726: }
6727: i=0;
6728: for (k=1; k<=(nlstate);k++){
6729: for (l=1; l<=(nlstate+ndeath);l++){
6730: i++;
6731: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6732: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6733: for (j=1; j<=i;j++){
6734: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6735: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6736: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6737: }
6738: }
6739: }/* end of loop for state */
6740: } /* end of loop for age */
6741: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6742: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6743: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6744: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6745:
6746: /* Confidence intervalle of pij */
6747: /*
6748: fprintf(ficgp,"\nunset parametric;unset label");
6749: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6750: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6751: 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);
6752: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6753: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6754: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6755: */
6756:
6757: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6758: first1=1;first2=2;
6759: for (k2=1; k2<=(nlstate);k2++){
6760: for (l2=1; l2<=(nlstate+ndeath);l2++){
6761: if(l2==k2) continue;
6762: j=(k2-1)*(nlstate+ndeath)+l2;
6763: for (k1=1; k1<=(nlstate);k1++){
6764: for (l1=1; l1<=(nlstate+ndeath);l1++){
6765: if(l1==k1) continue;
6766: i=(k1-1)*(nlstate+ndeath)+l1;
6767: if(i<=j) continue;
6768: for (age=bage; age<=fage; age ++){
6769: if ((int)age %5==0){
6770: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6771: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6772: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6773: mu1=mu[i][(int) age]/stepm*YEARM ;
6774: mu2=mu[j][(int) age]/stepm*YEARM;
6775: c12=cv12/sqrt(v1*v2);
6776: /* Computing eigen value of matrix of covariance */
6777: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6778: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6779: if ((lc2 <0) || (lc1 <0) ){
6780: if(first2==1){
6781: first1=0;
6782: 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);
6783: }
6784: 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);
6785: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6786: /* lc2=fabs(lc2); */
6787: }
1.220 brouard 6788:
1.222 brouard 6789: /* Eigen vectors */
1.280 brouard 6790: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6791: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6792: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6793: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6794: }else
6795: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6796: /*v21=sqrt(1.-v11*v11); *//* error */
6797: v21=(lc1-v1)/cv12*v11;
6798: v12=-v21;
6799: v22=v11;
6800: tnalp=v21/v11;
6801: if(first1==1){
6802: first1=0;
6803: 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);
6804: }
6805: 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);
6806: /*printf(fignu*/
6807: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6808: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6809: if(first==1){
6810: first=0;
6811: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6812: fprintf(ficgp,"\nset parametric;unset label");
6813: 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);
6814: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6815: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6816: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6817: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6818: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6819: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6820: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6821: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6822: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6823: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6824: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6825: 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", \
1.280 brouard 6826: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6827: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6828: }else{
6829: first=0;
6830: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6831: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6832: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6833: 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 6834: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6835: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6836: }/* if first */
6837: } /* age mod 5 */
6838: } /* end loop age */
6839: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6840: first=1;
6841: } /*l12 */
6842: } /* k12 */
6843: } /*l1 */
6844: }/* k1 */
6845: } /* loop on combination of covariates j1 */
6846: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6847: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6848: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6849: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6850: free_vector(xp,1,npar);
6851: fclose(ficresprob);
6852: fclose(ficresprobcov);
6853: fclose(ficresprobcor);
6854: fflush(ficgp);
6855: fflush(fichtmcov);
6856: }
1.126 brouard 6857:
6858:
6859: /******************* Printing html file ***********/
1.201 brouard 6860: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6861: int lastpass, int stepm, int weightopt, char model[],\
6862: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6863: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6864: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6865: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6866: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6867:
6868: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6869: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6870: </ul>");
1.237 brouard 6871: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6872: </ul>", model);
1.214 brouard 6873: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6874: 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",
6875: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6876: 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 6877: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6878: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6879: fprintf(fichtm,"\
6880: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6881: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6882: fprintf(fichtm,"\
1.217 brouard 6883: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6884: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6885: fprintf(fichtm,"\
1.288 brouard 6886: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6887: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6888: fprintf(fichtm,"\
1.288 brouard 6889: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6890: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6891: fprintf(fichtm,"\
1.211 brouard 6892: - (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 6893: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6894: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6895: if(prevfcast==1){
6896: fprintf(fichtm,"\
6897: - Prevalence projections by age and states: \
1.201 brouard 6898: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6899: }
1.126 brouard 6900:
6901:
1.225 brouard 6902: m=pow(2,cptcoveff);
1.222 brouard 6903: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6904:
1.264 brouard 6905: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6906:
6907: jj1=0;
6908:
6909: fprintf(fichtm," \n<ul>");
6910: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6911: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6912: if(m != 1 && TKresult[nres]!= k1)
6913: continue;
6914: jj1++;
6915: if (cptcovn > 0) {
6916: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6917: for (cpt=1; cpt<=cptcoveff;cpt++){
6918: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6919: }
6920: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6921: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6922: }
6923: fprintf(fichtm,"\">");
6924:
6925: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6926: fprintf(fichtm,"************ Results for covariates");
6927: for (cpt=1; cpt<=cptcoveff;cpt++){
6928: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6929: }
6930: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6931: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6932: }
6933: if(invalidvarcomb[k1]){
6934: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6935: continue;
6936: }
6937: fprintf(fichtm,"</a></li>");
6938: } /* cptcovn >0 */
6939: }
6940: fprintf(fichtm," \n</ul>");
6941:
1.222 brouard 6942: jj1=0;
1.237 brouard 6943:
6944: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6945: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6946: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6947: continue;
1.220 brouard 6948:
1.222 brouard 6949: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6950: jj1++;
6951: if (cptcovn > 0) {
1.264 brouard 6952: fprintf(fichtm,"\n<p><a name=\"rescov");
6953: for (cpt=1; cpt<=cptcoveff;cpt++){
6954: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6955: }
6956: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6957: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6958: }
6959: fprintf(fichtm,"\"</a>");
6960:
1.222 brouard 6961: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6962: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6963: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6964: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6965: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6966: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6967: }
1.237 brouard 6968: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6969: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6970: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6971: }
6972:
1.230 brouard 6973: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6974: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6975: if(invalidvarcomb[k1]){
6976: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6977: printf("\nCombination (%d) ignored because no cases \n",k1);
6978: continue;
6979: }
6980: }
6981: /* aij, bij */
1.259 brouard 6982: 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 6983: <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 6984: /* Pij */
1.241 brouard 6985: 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> \
6986: <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 6987: /* Quasi-incidences */
6988: 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 6989: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6990: 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 6991: 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> \
6992: <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 6993: /* Survival functions (period) in state j */
6994: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6995: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed 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> \
1.241 brouard 6996: <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 6997: }
6998: /* State specific survival functions (period) */
6999: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7000: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7001: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7002: <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 7003: }
1.288 brouard 7004: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7005: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7006: 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> \
7007: <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 7008: }
1.296 brouard 7009: if(prevbcast==1){
1.288 brouard 7010: /* Backward prevalence in each health state */
1.222 brouard 7011: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7012: 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 7013: <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 7014: }
1.217 brouard 7015: }
1.222 brouard 7016: if(prevfcast==1){
1.288 brouard 7017: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7018: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7019: 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) forward 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> \
1.296 brouard 7020: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7021: }
7022: }
1.296 brouard 7023: if(prevbcast==1){
1.268 brouard 7024: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7025: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7026: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7027: 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 \
7028: 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) \
7029: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7030: <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 7031: }
7032: }
1.220 brouard 7033:
1.222 brouard 7034: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7035: 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> \
7036: <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 7037: }
7038: /* } /\* end i1 *\/ */
7039: }/* End k1 */
7040: fprintf(fichtm,"</ul>");
1.126 brouard 7041:
1.222 brouard 7042: fprintf(fichtm,"\
1.126 brouard 7043: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7044: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7045: - 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 7046: But because parameters are usually highly correlated (a higher incidence of disability \
7047: and a higher incidence of recovery can give very close observed transition) it might \
7048: be very useful to look not only at linear confidence intervals estimated from the \
7049: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7050: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7051: covariance matrix of the one-step probabilities. \
7052: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7053:
1.222 brouard 7054: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7055: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7056: fprintf(fichtm,"\
1.126 brouard 7057: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7058: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7059:
1.222 brouard 7060: fprintf(fichtm,"\
1.126 brouard 7061: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7062: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7063: fprintf(fichtm,"\
1.126 brouard 7064: - 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): \
7065: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7066: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7067: fprintf(fichtm,"\
1.126 brouard 7068: - (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): \
7069: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7070: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7071: fprintf(fichtm,"\
1.288 brouard 7072: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the forward (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 7073: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7074: fprintf(fichtm,"\
1.128 brouard 7075: - 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 7076: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7077: fprintf(fichtm,"\
1.288 brouard 7078: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7079: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7080:
7081: /* if(popforecast==1) fprintf(fichtm,"\n */
7082: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7083: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7084: /* <br>",fileres,fileres,fileres,fileres); */
7085: /* else */
7086: /* 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 7087: fflush(fichtm);
7088: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7089:
1.225 brouard 7090: m=pow(2,cptcoveff);
1.222 brouard 7091: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7092:
1.222 brouard 7093: jj1=0;
1.237 brouard 7094:
1.241 brouard 7095: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7096: for(k1=1; k1<=m;k1++){
1.253 brouard 7097: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7098: continue;
1.222 brouard 7099: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7100: jj1++;
1.126 brouard 7101: if (cptcovn > 0) {
7102: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7103: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7104: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7105: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7106: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7107: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7108: }
7109:
1.126 brouard 7110: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7111:
1.222 brouard 7112: if(invalidvarcomb[k1]){
7113: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7114: continue;
7115: }
1.126 brouard 7116: }
7117: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7118: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7119: 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 7120: <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 7121: }
7122: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7123: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7124: true period expectancies (those weighted with period prevalences are also\
7125: drawn in addition to the population based expectancies computed using\
1.241 brouard 7126: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7127: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7128: /* } /\* end i1 *\/ */
7129: }/* End k1 */
1.241 brouard 7130: }/* End nres */
1.222 brouard 7131: fprintf(fichtm,"</ul>");
7132: fflush(fichtm);
1.126 brouard 7133: }
7134:
7135: /******************* Gnuplot file **************/
1.296 brouard 7136: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int prevbcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 7137:
7138: char dirfileres[132],optfileres[132];
1.264 brouard 7139: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7140: 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 7141: int lv=0, vlv=0, kl=0;
1.130 brouard 7142: int ng=0;
1.201 brouard 7143: int vpopbased;
1.223 brouard 7144: int ioffset; /* variable offset for columns */
1.270 brouard 7145: int iyearc=1; /* variable column for year of projection */
7146: int iagec=1; /* variable column for age of projection */
1.235 brouard 7147: int nres=0; /* Index of resultline */
1.266 brouard 7148: int istart=1; /* For starting graphs in projections */
1.219 brouard 7149:
1.126 brouard 7150: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7151: /* printf("Problem with file %s",optionfilegnuplot); */
7152: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7153: /* } */
7154:
7155: /*#ifdef windows */
7156: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7157: /*#endif */
1.225 brouard 7158: m=pow(2,cptcoveff);
1.126 brouard 7159:
1.274 brouard 7160: /* diagram of the model */
7161: fprintf(ficgp,"\n#Diagram of the model \n");
7162: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7163: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7164: fprintf(ficgp,"\n#Peripheral arrows\nset for [i=1:%d] for [j=1:%d] arrow i*10+j from cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.95*(cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) - cos(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta2:0)), -0.95*(sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) - sin(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d))+( i!=j?(i-j)/abs(i-j)*delta2:0)) ls (i < j? 1:2)\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
7165:
7166: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
7167: fprintf(ficgp,"\n#show arrow\nunset label\n");
7168: fprintf(ficgp,"\n#States labels, starting from 2 (2-i) instead of (1-i), was (i-1)\nset for [i=1:%d] label i sprintf(\"State %%d\",i) center at cos(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)), yoff+sin(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)) font \"helvetica, 16\" tc rgbcolor \"blue\"\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
7169: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7170: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7171: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7172: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7173:
1.202 brouard 7174: /* Contribution to likelihood */
7175: /* Plot the probability implied in the likelihood */
1.223 brouard 7176: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7177: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7178: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7179: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7180: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7181: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7182: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7183: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7184: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7185: 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));
7186: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7187: 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));
7188: for (i=1; i<= nlstate ; i ++) {
7189: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7190: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7191: 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);
7192: for (j=2; j<= nlstate+ndeath ; j ++) {
7193: 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);
7194: }
7195: fprintf(ficgp,";\nset out; unset ylabel;\n");
7196: }
7197: /* 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 */
7198: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7199: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7200: fprintf(ficgp,"\nset out;unset log\n");
7201: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7202:
1.126 brouard 7203: strcpy(dirfileres,optionfilefiname);
7204: strcpy(optfileres,"vpl");
1.223 brouard 7205: /* 1eme*/
1.238 brouard 7206: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7207: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7208: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7209: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7210: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7211: continue;
7212: /* We are interested in selected combination by the resultline */
1.246 brouard 7213: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7214: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7215: strcpy(gplotlabel,"(");
1.238 brouard 7216: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7217: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7218: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7219: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7220: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7221: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7222: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7223: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7224: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7225: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7226: }
7227: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7228: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7229: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7230: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7231: }
7232: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7233: /* printf("\n#\n"); */
1.238 brouard 7234: fprintf(ficgp,"\n#\n");
7235: if(invalidvarcomb[k1]){
1.260 brouard 7236: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7237: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7238: continue;
7239: }
1.235 brouard 7240:
1.241 brouard 7241: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7242: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7243: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7244: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7245: 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);
7246: /* 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); */
7247: /* k1-1 error should be nres-1*/
1.238 brouard 7248: for (i=1; i<= nlstate ; i ++) {
7249: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7250: else fprintf(ficgp," %%*lf (%%*lf)");
7251: }
1.288 brouard 7252: fprintf(ficgp,"\" t\"Forward 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 7253: for (i=1; i<= nlstate ; i ++) {
7254: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7255: else fprintf(ficgp," %%*lf (%%*lf)");
7256: }
1.260 brouard 7257: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7258: for (i=1; i<= nlstate ; i ++) {
7259: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7260: else fprintf(ficgp," %%*lf (%%*lf)");
7261: }
1.265 brouard 7262: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7263:
7264: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7265: if(cptcoveff ==0){
1.271 brouard 7266: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7267: }else{
7268: kl=0;
7269: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7270: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7271: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7272: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7273: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7274: vlv= nbcode[Tvaraff[k]][lv];
7275: kl++;
7276: /* 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 *\/ */
7277: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7278: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7279: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7280: if(k==cptcoveff){
7281: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7282: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7283: }else{
7284: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7285: kl++;
7286: }
7287: } /* end covariate */
7288: } /* end if no covariate */
7289:
1.296 brouard 7290: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7291: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7292: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7293: if(cptcoveff ==0){
1.245 brouard 7294: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7295: }else{
7296: kl=0;
7297: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7298: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7299: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7300: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7301: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7302: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7303: kl++;
1.238 brouard 7304: /* 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 *\/ */
7305: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7306: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7307: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7308: if(k==cptcoveff){
1.245 brouard 7309: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7310: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7311: }else{
7312: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7313: kl++;
7314: }
7315: } /* end covariate */
7316: } /* end if no covariate */
1.296 brouard 7317: if(prevbcast == 1){
1.268 brouard 7318: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7319: /* k1-1 error should be nres-1*/
7320: for (i=1; i<= nlstate ; i ++) {
7321: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7322: else fprintf(ficgp," %%*lf (%%*lf)");
7323: }
1.271 brouard 7324: 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 7325: for (i=1; i<= nlstate ; i ++) {
7326: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7327: else fprintf(ficgp," %%*lf (%%*lf)");
7328: }
1.276 brouard 7329: fprintf(ficgp,"\" t\"95%% CI\" w l lt 4,\"%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 7330: for (i=1; i<= nlstate ; i ++) {
7331: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7332: else fprintf(ficgp," %%*lf (%%*lf)");
7333: }
1.274 brouard 7334: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7335: } /* end if backprojcast */
1.296 brouard 7336: } /* end if prevbcast */
1.276 brouard 7337: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7338: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7339: } /* nres */
1.201 brouard 7340: } /* k1 */
7341: } /* cpt */
1.235 brouard 7342:
7343:
1.126 brouard 7344: /*2 eme*/
1.238 brouard 7345: for (k1=1; k1<= m ; k1 ++){
7346: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7347: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7348: continue;
7349: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7350: strcpy(gplotlabel,"(");
1.238 brouard 7351: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7352: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7353: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7354: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7355: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7356: vlv= nbcode[Tvaraff[k]][lv];
7357: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7358: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7359: }
1.237 brouard 7360: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7361: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7362: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7363: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7364: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7365: }
1.264 brouard 7366: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7367: fprintf(ficgp,"\n#\n");
1.223 brouard 7368: if(invalidvarcomb[k1]){
7369: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7370: continue;
7371: }
1.219 brouard 7372:
1.241 brouard 7373: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7374: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7375: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7376: if(vpopbased==0){
1.238 brouard 7377: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7378: }else
1.238 brouard 7379: fprintf(ficgp,"\nreplot ");
7380: for (i=1; i<= nlstate+1 ; i ++) {
7381: k=2*i;
1.261 brouard 7382: 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 7383: for (j=1; j<= nlstate+1 ; j ++) {
7384: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7385: else fprintf(ficgp," %%*lf (%%*lf)");
7386: }
7387: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7388: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7389: 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 7390: for (j=1; j<= nlstate+1 ; j ++) {
7391: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7392: else fprintf(ficgp," %%*lf (%%*lf)");
7393: }
7394: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7395: 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 7396: for (j=1; j<= nlstate+1 ; j ++) {
7397: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7398: else fprintf(ficgp," %%*lf (%%*lf)");
7399: }
7400: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7401: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7402: } /* state */
7403: } /* vpopbased */
1.264 brouard 7404: 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 7405: } /* end nres */
7406: } /* k1 end 2 eme*/
7407:
7408:
7409: /*3eme*/
7410: for (k1=1; k1<= m ; k1 ++){
7411: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7412: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7413: continue;
7414:
7415: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7416: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7417: strcpy(gplotlabel,"(");
1.238 brouard 7418: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7419: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7420: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7421: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7422: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7423: vlv= nbcode[Tvaraff[k]][lv];
7424: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7425: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7426: }
7427: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7428: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7429: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7430: }
1.264 brouard 7431: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7432: fprintf(ficgp,"\n#\n");
7433: if(invalidvarcomb[k1]){
7434: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7435: continue;
7436: }
7437:
7438: /* k=2+nlstate*(2*cpt-2); */
7439: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7440: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7441: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7442: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7443: 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 7444: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7445: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7446: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7447: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7448: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7449: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7450:
1.238 brouard 7451: */
7452: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7453: 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 7454: /* 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 7455:
1.238 brouard 7456: }
1.261 brouard 7457: 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 7458: }
1.264 brouard 7459: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7460: } /* end nres */
7461: } /* end kl 3eme */
1.126 brouard 7462:
1.223 brouard 7463: /* 4eme */
1.201 brouard 7464: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7465: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7466: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7467: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7468: continue;
1.238 brouard 7469: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7470: strcpy(gplotlabel,"(");
1.238 brouard 7471: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7472: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7473: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7474: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7475: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7476: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7477: vlv= nbcode[Tvaraff[k]][lv];
7478: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7479: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7480: }
7481: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7482: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7483: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7484: }
1.264 brouard 7485: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7486: fprintf(ficgp,"\n#\n");
7487: if(invalidvarcomb[k1]){
7488: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7489: continue;
1.223 brouard 7490: }
1.238 brouard 7491:
1.241 brouard 7492: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7493: 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 7494: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7495: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7496: k=3;
7497: for (i=1; i<= nlstate ; i ++){
7498: if(i==1){
7499: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7500: }else{
7501: fprintf(ficgp,", '' ");
7502: }
7503: l=(nlstate+ndeath)*(i-1)+1;
7504: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7505: for (j=2; j<= nlstate+ndeath ; j ++)
7506: fprintf(ficgp,"+$%d",k+l+j-1);
7507: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7508: } /* nlstate */
1.264 brouard 7509: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7510: } /* end cpt state*/
7511: } /* end nres */
7512: } /* end covariate k1 */
7513:
1.220 brouard 7514: /* 5eme */
1.201 brouard 7515: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7516: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7517: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7518: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7519: continue;
1.238 brouard 7520: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7521: strcpy(gplotlabel,"(");
1.238 brouard 7522: 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);
7523: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7524: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7525: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7526: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7527: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7528: vlv= nbcode[Tvaraff[k]][lv];
7529: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7530: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7531: }
7532: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7533: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7534: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7535: }
1.264 brouard 7536: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7537: fprintf(ficgp,"\n#\n");
7538: if(invalidvarcomb[k1]){
7539: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7540: continue;
7541: }
1.227 brouard 7542:
1.241 brouard 7543: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7544: 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 7545: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7546: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7547: k=3;
7548: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7549: if(j==1)
7550: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7551: else
7552: fprintf(ficgp,", '' ");
7553: l=(nlstate+ndeath)*(cpt-1) +j;
7554: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7555: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7556: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7557: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7558: } /* nlstate */
7559: fprintf(ficgp,", '' ");
7560: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7561: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7562: l=(nlstate+ndeath)*(cpt-1) +j;
7563: if(j < nlstate)
7564: fprintf(ficgp,"$%d +",k+l);
7565: else
7566: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7567: }
1.264 brouard 7568: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7569: } /* end cpt state*/
7570: } /* end covariate */
7571: } /* end nres */
1.227 brouard 7572:
1.220 brouard 7573: /* 6eme */
1.202 brouard 7574: /* CV preval stable (period) for each covariate */
1.237 brouard 7575: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7576: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7577: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7578: continue;
1.255 brouard 7579: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7580: strcpy(gplotlabel,"(");
1.288 brouard 7581: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7582: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7583: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7584: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7585: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7586: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7587: vlv= nbcode[Tvaraff[k]][lv];
7588: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7589: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7590: }
1.237 brouard 7591: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7592: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7593: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7594: }
1.264 brouard 7595: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7596: fprintf(ficgp,"\n#\n");
1.223 brouard 7597: if(invalidvarcomb[k1]){
1.227 brouard 7598: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7599: continue;
1.223 brouard 7600: }
1.227 brouard 7601:
1.241 brouard 7602: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7603: 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 7604: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7605: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7606: k=3; /* Offset */
1.255 brouard 7607: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7608: if(i==1)
7609: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7610: else
7611: fprintf(ficgp,", '' ");
1.255 brouard 7612: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7613: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7614: for (j=2; j<= nlstate ; j ++)
7615: fprintf(ficgp,"+$%d",k+l+j-1);
7616: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7617: } /* nlstate */
1.264 brouard 7618: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7619: } /* end cpt state*/
7620: } /* end covariate */
1.227 brouard 7621:
7622:
1.220 brouard 7623: /* 7eme */
1.296 brouard 7624: if(prevbcast == 1){
1.288 brouard 7625: /* CV backward prevalence for each covariate */
1.237 brouard 7626: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7627: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7628: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7629: continue;
1.268 brouard 7630: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7631: strcpy(gplotlabel,"(");
1.288 brouard 7632: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7633: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7634: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7635: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7636: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7637: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7638: vlv= nbcode[Tvaraff[k]][lv];
7639: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7640: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7641: }
1.237 brouard 7642: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7643: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7644: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7645: }
1.264 brouard 7646: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7647: fprintf(ficgp,"\n#\n");
7648: if(invalidvarcomb[k1]){
7649: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7650: continue;
7651: }
7652:
1.241 brouard 7653: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7654: 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 7655: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7656: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7657: k=3; /* Offset */
1.268 brouard 7658: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7659: if(i==1)
7660: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7661: else
7662: fprintf(ficgp,", '' ");
7663: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7664: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7665: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7666: /* 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 7667: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7668: /* for (j=2; j<= nlstate ; j ++) */
7669: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7670: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7671: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7672: } /* nlstate */
1.264 brouard 7673: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7674: } /* end cpt state*/
7675: } /* end covariate */
1.296 brouard 7676: } /* End if prevbcast */
1.218 brouard 7677:
1.223 brouard 7678: /* 8eme */
1.218 brouard 7679: if(prevfcast==1){
1.288 brouard 7680: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7681:
1.237 brouard 7682: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7683: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7684: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7685: continue;
1.211 brouard 7686: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7687: strcpy(gplotlabel,"(");
1.288 brouard 7688: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7689: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7690: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7691: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7692: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7693: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7694: vlv= nbcode[Tvaraff[k]][lv];
7695: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7696: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7697: }
1.237 brouard 7698: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7699: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7700: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7701: }
1.264 brouard 7702: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7703: fprintf(ficgp,"\n#\n");
7704: if(invalidvarcomb[k1]){
7705: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7706: continue;
7707: }
7708:
7709: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7710: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7711: 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 7712: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7713: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7714:
7715: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7716: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7717: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7718: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7719: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7720: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7721: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7722: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7723: if(i==istart){
1.227 brouard 7724: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7725: }else{
7726: fprintf(ficgp,",\\\n '' ");
7727: }
7728: if(cptcoveff ==0){ /* No covariate */
7729: ioffset=2; /* Age is in 2 */
7730: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7731: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7732: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7733: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7734: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7735: if(i==nlstate+1){
1.270 brouard 7736: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7737: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7738: fprintf(ficgp,",\\\n '' ");
7739: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7740: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7741: offyear, \
1.268 brouard 7742: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7743: }else
1.227 brouard 7744: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7745: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7746: }else{ /* more than 2 covariates */
1.270 brouard 7747: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7748: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7749: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7750: iyearc=ioffset-1;
7751: iagec=ioffset;
1.227 brouard 7752: fprintf(ficgp," u %d:(",ioffset);
7753: kl=0;
7754: strcpy(gplotcondition,"(");
7755: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7756: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7757: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7758: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7759: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7760: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7761: kl++;
7762: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7763: kl++;
7764: if(k <cptcoveff && cptcoveff>1)
7765: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7766: }
7767: strcpy(gplotcondition+strlen(gplotcondition),")");
7768: /* 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 *\/ */
7769: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7770: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7771: /* '' 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*/
7772: if(i==nlstate+1){
1.270 brouard 7773: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7774: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7775: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7776: fprintf(ficgp," u %d:(",iagec);
7777: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7778: iyearc, iagec, offyear, \
7779: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7780: /* '' 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 7781: }else{
7782: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7783: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7784: }
7785: } /* end if covariate */
7786: } /* nlstate */
1.264 brouard 7787: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7788: } /* end cpt state*/
7789: } /* end covariate */
7790: } /* End if prevfcast */
1.227 brouard 7791:
1.296 brouard 7792: if(prevbcast==1){
1.268 brouard 7793: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7794:
7795: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7796: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7797: if(m != 1 && TKresult[nres]!= k1)
7798: continue;
7799: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7800: strcpy(gplotlabel,"(");
7801: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7802: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7803: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7804: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7805: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7806: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7807: vlv= nbcode[Tvaraff[k]][lv];
7808: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7809: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7810: }
7811: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7812: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7813: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7814: }
7815: strcpy(gplotlabel+strlen(gplotlabel),")");
7816: fprintf(ficgp,"\n#\n");
7817: if(invalidvarcomb[k1]){
7818: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7819: continue;
7820: }
7821:
7822: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7823: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7824: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7825: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7826: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7827:
7828: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7829: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7830: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7831: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7832: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7833: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7834: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7835: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7836: if(i==istart){
7837: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7838: }else{
7839: fprintf(ficgp,",\\\n '' ");
7840: }
7841: if(cptcoveff ==0){ /* No covariate */
7842: ioffset=2; /* Age is in 2 */
7843: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7844: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7845: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7846: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7847: fprintf(ficgp," u %d:(", ioffset);
7848: if(i==nlstate+1){
1.270 brouard 7849: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7850: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7851: fprintf(ficgp,",\\\n '' ");
7852: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7853: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7854: offbyear, \
7855: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7856: }else
7857: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7858: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7859: }else{ /* more than 2 covariates */
1.270 brouard 7860: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7861: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7862: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7863: iyearc=ioffset-1;
7864: iagec=ioffset;
1.268 brouard 7865: fprintf(ficgp," u %d:(",ioffset);
7866: kl=0;
7867: strcpy(gplotcondition,"(");
7868: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7869: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7870: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7871: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7872: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7873: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7874: kl++;
7875: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7876: kl++;
7877: if(k <cptcoveff && cptcoveff>1)
7878: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7879: }
7880: strcpy(gplotcondition+strlen(gplotcondition),")");
7881: /* 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 *\/ */
7882: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7883: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7884: /* '' 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*/
7885: if(i==nlstate+1){
1.270 brouard 7886: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7887: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7888: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7889: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7890: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7891: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7892: iyearc,iagec,offbyear, \
7893: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7894: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7895: }else{
7896: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7897: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7898: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7899: }
7900: } /* end if covariate */
7901: } /* nlstate */
7902: fprintf(ficgp,"\nset out; unset label;\n");
7903: } /* end cpt state*/
7904: } /* end covariate */
1.296 brouard 7905: } /* End if prevbcast */
1.268 brouard 7906:
1.227 brouard 7907:
1.238 brouard 7908: /* 9eme writing MLE parameters */
7909: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7910: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7911: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7912: for(k=1; k <=(nlstate+ndeath); k++){
7913: if (k != i) {
1.227 brouard 7914: fprintf(ficgp,"# current state %d\n",k);
7915: for(j=1; j <=ncovmodel; j++){
7916: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7917: jk++;
7918: }
7919: fprintf(ficgp,"\n");
1.126 brouard 7920: }
7921: }
1.223 brouard 7922: }
1.187 brouard 7923: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7924:
1.145 brouard 7925: /*goto avoid;*/
1.238 brouard 7926: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7927: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7928: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7929: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7930: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7931: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7932: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7933: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7934: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7935: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7936: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7937: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7938: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7939: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7940: fprintf(ficgp,"#\n");
1.223 brouard 7941: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7942: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7943: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7944: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7945: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7946: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7947: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7948: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7949: continue;
1.264 brouard 7950: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7951: strcpy(gplotlabel,"(");
1.276 brouard 7952: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7953: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7954: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7955: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7956: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7957: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7958: vlv= nbcode[Tvaraff[k]][lv];
7959: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7960: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7961: }
1.237 brouard 7962: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7963: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7964: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7965: }
1.264 brouard 7966: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7967: fprintf(ficgp,"\n#\n");
1.264 brouard 7968: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7969: fprintf(ficgp,"\nset key outside ");
7970: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7971: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7972: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7973: if (ng==1){
7974: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7975: fprintf(ficgp,"\nunset log y");
7976: }else if (ng==2){
7977: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7978: fprintf(ficgp,"\nset log y");
7979: }else if (ng==3){
7980: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7981: fprintf(ficgp,"\nset log y");
7982: }else
7983: fprintf(ficgp,"\nunset title ");
7984: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7985: i=1;
7986: for(k2=1; k2<=nlstate; k2++) {
7987: k3=i;
7988: for(k=1; k<=(nlstate+ndeath); k++) {
7989: if (k != k2){
7990: switch( ng) {
7991: case 1:
7992: if(nagesqr==0)
7993: fprintf(ficgp," p%d+p%d*x",i,i+1);
7994: else /* nagesqr =1 */
7995: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7996: break;
7997: case 2: /* ng=2 */
7998: if(nagesqr==0)
7999: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8000: else /* nagesqr =1 */
8001: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8002: break;
8003: case 3:
8004: if(nagesqr==0)
8005: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8006: else /* nagesqr =1 */
8007: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8008: break;
8009: }
8010: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8011: ijp=1; /* product no age */
8012: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8013: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8014: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8015: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8016: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8017: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8018: if(DummyV[j]==0){
8019: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8020: }else{ /* quantitative */
8021: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8022: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8023: }
8024: ij++;
1.237 brouard 8025: }
1.268 brouard 8026: }
8027: }else if(cptcovprod >0){
8028: if(j==Tprod[ijp]) { /* */
8029: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8030: if(ijp <=cptcovprod) { /* Product */
8031: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8032: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8033: /* 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)]); */
8034: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8035: }else{ /* Vn is dummy and Vm is quanti */
8036: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8037: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8038: }
8039: }else{ /* Vn*Vm Vn is quanti */
8040: if(DummyV[Tvard[ijp][2]]==0){
8041: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8042: }else{ /* Both quanti */
8043: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8044: }
1.237 brouard 8045: }
1.268 brouard 8046: ijp++;
1.237 brouard 8047: }
1.268 brouard 8048: } /* end Tprod */
1.237 brouard 8049: } else{ /* simple covariate */
1.264 brouard 8050: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8051: if(Dummy[j]==0){
8052: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8053: }else{ /* quantitative */
8054: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8055: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8056: }
1.237 brouard 8057: } /* end simple */
8058: } /* end j */
1.223 brouard 8059: }else{
8060: i=i-ncovmodel;
8061: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8062: fprintf(ficgp," (1.");
8063: }
1.227 brouard 8064:
1.223 brouard 8065: if(ng != 1){
8066: fprintf(ficgp,")/(1");
1.227 brouard 8067:
1.264 brouard 8068: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8069: if(nagesqr==0)
1.264 brouard 8070: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8071: else /* nagesqr =1 */
1.264 brouard 8072: 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 8073:
1.223 brouard 8074: ij=1;
8075: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8076: if(cptcovage >0){
8077: if((j-2)==Tage[ij]) { /* Bug valgrind */
8078: if(ij <=cptcovage) { /* Bug valgrind */
8079: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8080: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8081: ij++;
8082: }
8083: }
8084: }else
8085: 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 8086: }
8087: fprintf(ficgp,")");
8088: }
8089: fprintf(ficgp,")");
8090: if(ng ==2)
1.276 brouard 8091: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"p%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 8092: else /* ng= 3 */
1.276 brouard 8093: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 8094: }else{ /* end ng <> 1 */
8095: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8096: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"logit(p%d%d)\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 8097: }
8098: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8099: fprintf(ficgp,",");
8100: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8101: fprintf(ficgp,",");
8102: i=i+ncovmodel;
8103: } /* end k */
8104: } /* end k2 */
1.276 brouard 8105: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8106: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8107: } /* end k1 */
1.223 brouard 8108: } /* end ng */
8109: /* avoid: */
8110: fflush(ficgp);
1.126 brouard 8111: } /* end gnuplot */
8112:
8113:
8114: /*************** Moving average **************/
1.219 brouard 8115: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8116: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8117:
1.222 brouard 8118: int i, cpt, cptcod;
8119: int modcovmax =1;
8120: int mobilavrange, mob;
8121: int iage=0;
1.288 brouard 8122: int firstA1=0, firstA2=0;
1.222 brouard 8123:
1.266 brouard 8124: double sum=0., sumr=0.;
1.222 brouard 8125: double age;
1.266 brouard 8126: double *sumnewp, *sumnewm, *sumnewmr;
8127: double *agemingood, *agemaxgood;
8128: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8129:
8130:
1.278 brouard 8131: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8132: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8133:
8134: sumnewp = vector(1,ncovcombmax);
8135: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8136: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8137: agemingood = vector(1,ncovcombmax);
1.266 brouard 8138: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8139: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8140: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8141:
8142: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8143: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8144: sumnewp[cptcod]=0.;
1.266 brouard 8145: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8146: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8147: }
8148: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8149:
1.266 brouard 8150: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8151: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8152: else mobilavrange=mobilav;
8153: for (age=bage; age<=fage; age++)
8154: for (i=1; i<=nlstate;i++)
8155: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8156: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8157: /* We keep the original values on the extreme ages bage, fage and for
8158: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8159: we use a 5 terms etc. until the borders are no more concerned.
8160: */
8161: for (mob=3;mob <=mobilavrange;mob=mob+2){
8162: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8163: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8164: sumnewm[cptcod]=0.;
8165: for (i=1; i<=nlstate;i++){
1.222 brouard 8166: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8167: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8168: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8169: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8170: }
8171: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8172: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8173: } /* end i */
8174: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8175: } /* end cptcod */
1.222 brouard 8176: }/* end age */
8177: }/* end mob */
1.266 brouard 8178: }else{
8179: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8180: return -1;
1.266 brouard 8181: }
8182:
8183: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8184: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8185: if(invalidvarcomb[cptcod]){
8186: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8187: continue;
8188: }
1.219 brouard 8189:
1.266 brouard 8190: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8191: sumnewm[cptcod]=0.;
8192: sumnewmr[cptcod]=0.;
8193: for (i=1; i<=nlstate;i++){
8194: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8195: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8196: }
8197: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8198: agemingoodr[cptcod]=age;
8199: }
8200: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8201: agemingood[cptcod]=age;
8202: }
8203: } /* age */
8204: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8205: sumnewm[cptcod]=0.;
1.266 brouard 8206: sumnewmr[cptcod]=0.;
1.222 brouard 8207: for (i=1; i<=nlstate;i++){
8208: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8209: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8210: }
8211: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8212: agemaxgoodr[cptcod]=age;
1.222 brouard 8213: }
8214: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8215: agemaxgood[cptcod]=age;
8216: }
8217: } /* age */
8218: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8219: /* but they will change */
1.288 brouard 8220: firstA1=0;firstA2=0;
1.266 brouard 8221: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8222: sumnewm[cptcod]=0.;
8223: sumnewmr[cptcod]=0.;
8224: for (i=1; i<=nlstate;i++){
8225: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8226: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8227: }
8228: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8229: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8230: agemaxgoodr[cptcod]=age; /* age min */
8231: for (i=1; i<=nlstate;i++)
8232: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8233: }else{ /* bad we change the value with the values of good ages */
8234: for (i=1; i<=nlstate;i++){
8235: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8236: } /* i */
8237: } /* end bad */
8238: }else{
8239: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8240: agemaxgood[cptcod]=age;
8241: }else{ /* bad we change the value with the values of good ages */
8242: for (i=1; i<=nlstate;i++){
8243: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8244: } /* i */
8245: } /* end bad */
8246: }/* end else */
8247: sum=0.;sumr=0.;
8248: for (i=1; i<=nlstate;i++){
8249: sum+=mobaverage[(int)age][i][cptcod];
8250: sumr+=probs[(int)age][i][cptcod];
8251: }
8252: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8253: if(!firstA1){
8254: firstA1=1;
8255: 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. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
8256: }
8257: fprintf(ficlog,"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 8258: } /* end bad */
8259: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8260: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8261: if(!firstA2){
8262: firstA2=1;
8263: 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. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
8264: }
8265: fprintf(ficlog,"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 8266: } /* end bad */
8267: }/* age */
1.266 brouard 8268:
8269: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8270: sumnewm[cptcod]=0.;
1.266 brouard 8271: sumnewmr[cptcod]=0.;
1.222 brouard 8272: for (i=1; i<=nlstate;i++){
8273: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8274: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8275: }
8276: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8277: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8278: agemingoodr[cptcod]=age;
8279: for (i=1; i<=nlstate;i++)
8280: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8281: }else{ /* bad we change the value with the values of good ages */
8282: for (i=1; i<=nlstate;i++){
8283: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8284: } /* i */
8285: } /* end bad */
8286: }else{
8287: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8288: agemingood[cptcod]=age;
8289: }else{ /* bad */
8290: for (i=1; i<=nlstate;i++){
8291: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8292: } /* i */
8293: } /* end bad */
8294: }/* end else */
8295: sum=0.;sumr=0.;
8296: for (i=1; i<=nlstate;i++){
8297: sum+=mobaverage[(int)age][i][cptcod];
8298: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8299: }
1.266 brouard 8300: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8301: 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 8302: } /* end bad */
8303: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8304: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8305: 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 8306: } /* end bad */
8307: }/* age */
1.266 brouard 8308:
1.222 brouard 8309:
8310: for (age=bage; age<=fage; age++){
1.235 brouard 8311: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8312: sumnewp[cptcod]=0.;
8313: sumnewm[cptcod]=0.;
8314: for (i=1; i<=nlstate;i++){
8315: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8316: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8317: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8318: }
8319: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8320: }
8321: /* printf("\n"); */
8322: /* } */
1.266 brouard 8323:
1.222 brouard 8324: /* brutal averaging */
1.266 brouard 8325: /* for (i=1; i<=nlstate;i++){ */
8326: /* for (age=1; age<=bage; age++){ */
8327: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8328: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8329: /* } */
8330: /* for (age=fage; age<=AGESUP; age++){ */
8331: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8332: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8333: /* } */
8334: /* } /\* end i status *\/ */
8335: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8336: /* for (age=1; age<=AGESUP; age++){ */
8337: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8338: /* mobaverage[(int)age][i][cptcod]=0.; */
8339: /* } */
8340: /* } */
1.222 brouard 8341: }/* end cptcod */
1.266 brouard 8342: free_vector(agemaxgoodr,1, ncovcombmax);
8343: free_vector(agemaxgood,1, ncovcombmax);
8344: free_vector(agemingood,1, ncovcombmax);
8345: free_vector(agemingoodr,1, ncovcombmax);
8346: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8347: free_vector(sumnewm,1, ncovcombmax);
8348: free_vector(sumnewp,1, ncovcombmax);
8349: return 0;
8350: }/* End movingaverage */
1.218 brouard 8351:
1.126 brouard 8352:
1.296 brouard 8353:
1.126 brouard 8354: /************** Forecasting ******************/
1.296 brouard 8355: /* void prevforecast(char fileres[], double dateintmean, double anprojd, double mprojd, double jprojd, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anprojf, double p[], int cptcoveff)*/
8356: void prevforecast(char fileres[], double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
8357: /* dateintemean, mean date of interviews
8358: dateprojd, year, month, day of starting projection
8359: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8360: agemin, agemax range of age
8361: dateprev1 dateprev2 range of dates during which prevalence is computed
8362: */
1.296 brouard 8363: /* double anprojd, mprojd, jprojd; */
8364: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8365: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8366: double agec; /* generic age */
1.296 brouard 8367: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8368: double *popeffectif,*popcount;
8369: double ***p3mat;
1.218 brouard 8370: /* double ***mobaverage; */
1.126 brouard 8371: char fileresf[FILENAMELENGTH];
8372:
8373: agelim=AGESUP;
1.211 brouard 8374: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8375: in each health status at the date of interview (if between dateprev1 and dateprev2).
8376: We still use firstpass and lastpass as another selection.
8377: */
1.214 brouard 8378: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8379: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8380:
1.201 brouard 8381: strcpy(fileresf,"F_");
8382: strcat(fileresf,fileresu);
1.126 brouard 8383: if((ficresf=fopen(fileresf,"w"))==NULL) {
8384: printf("Problem with forecast resultfile: %s\n", fileresf);
8385: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8386: }
1.235 brouard 8387: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8388: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8389:
1.225 brouard 8390: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8391:
8392:
8393: stepsize=(int) (stepm+YEARM-1)/YEARM;
8394: if (stepm<=12) stepsize=1;
8395: if(estepm < stepm){
8396: printf ("Problem %d lower than %d\n",estepm, stepm);
8397: }
1.270 brouard 8398: else{
8399: hstepm=estepm;
8400: }
8401: if(estepm > stepm){ /* Yes every two year */
8402: stepsize=2;
8403: }
1.296 brouard 8404: hstepm=hstepm/stepm;
1.126 brouard 8405:
1.296 brouard 8406:
8407: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8408: /* fractional in yp1 *\/ */
8409: /* aintmean=yp; */
8410: /* yp2=modf((yp1*12),&yp); */
8411: /* mintmean=yp; */
8412: /* yp1=modf((yp2*30.5),&yp); */
8413: /* jintmean=yp; */
8414: /* if(jintmean==0) jintmean=1; */
8415: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8416:
1.296 brouard 8417:
8418: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8419: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8420: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8421: i1=pow(2,cptcoveff);
1.126 brouard 8422: if (cptcovn < 1){i1=1;}
8423:
1.296 brouard 8424: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8425:
8426: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8427:
1.126 brouard 8428: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8429: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8430: for(k=1; k<=i1;k++){
1.253 brouard 8431: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8432: continue;
1.227 brouard 8433: if(invalidvarcomb[k]){
8434: printf("\nCombination (%d) projection ignored because no cases \n",k);
8435: continue;
8436: }
8437: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8438: for(j=1;j<=cptcoveff;j++) {
8439: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8440: }
1.235 brouard 8441: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8442: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8443: }
1.227 brouard 8444: fprintf(ficresf," yearproj age");
8445: for(j=1; j<=nlstate+ndeath;j++){
8446: for(i=1; i<=nlstate;i++)
8447: fprintf(ficresf," p%d%d",i,j);
8448: fprintf(ficresf," wp.%d",j);
8449: }
1.296 brouard 8450: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8451: fprintf(ficresf,"\n");
1.296 brouard 8452: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8453: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8454: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8455: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8456: nhstepm = nhstepm/hstepm;
8457: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8458: oldm=oldms;savm=savms;
1.268 brouard 8459: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8460: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8461: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8462: for (h=0; h<=nhstepm; h++){
8463: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8464: break;
8465: }
8466: }
8467: fprintf(ficresf,"\n");
8468: for(j=1;j<=cptcoveff;j++)
8469: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8470: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8471:
8472: for(j=1; j<=nlstate+ndeath;j++) {
8473: ppij=0.;
8474: for(i=1; i<=nlstate;i++) {
1.278 brouard 8475: if (mobilav>=1)
8476: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8477: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8478: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8479: }
1.268 brouard 8480: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8481: } /* end i */
8482: fprintf(ficresf," %.3f", ppij);
8483: }/* end j */
1.227 brouard 8484: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8485: } /* end agec */
1.266 brouard 8486: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8487: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8488: } /* end yearp */
8489: } /* end k */
1.219 brouard 8490:
1.126 brouard 8491: fclose(ficresf);
1.215 brouard 8492: printf("End of Computing forecasting \n");
8493: fprintf(ficlog,"End of Computing forecasting\n");
8494:
1.126 brouard 8495: }
8496:
1.269 brouard 8497: /************** Back Forecasting ******************/
1.296 brouard 8498: /* 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){ */
8499: void prevbackforecast(char fileres[], double ***prevacurrent, double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
8500: /* back1, year, month, day of starting backprojection
1.267 brouard 8501: agemin, agemax range of age
8502: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8503: anback2 year of end of backprojection (same day and month as back1).
8504: prevacurrent and prev are prevalences.
1.267 brouard 8505: */
8506: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8507: double agec; /* generic age */
1.302 ! brouard 8508: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8509: double *popeffectif,*popcount;
8510: double ***p3mat;
8511: /* double ***mobaverage; */
8512: char fileresfb[FILENAMELENGTH];
8513:
1.268 brouard 8514: agelim=AGEINF;
1.267 brouard 8515: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8516: in each health status at the date of interview (if between dateprev1 and dateprev2).
8517: We still use firstpass and lastpass as another selection.
8518: */
8519: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8520: /* firstpass, lastpass, stepm, weightopt, model); */
8521:
8522: /*Do we need to compute prevalence again?*/
8523:
8524: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8525:
8526: strcpy(fileresfb,"FB_");
8527: strcat(fileresfb,fileresu);
8528: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8529: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8530: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8531: }
8532: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8533: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8534:
8535: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8536:
8537:
8538: stepsize=(int) (stepm+YEARM-1)/YEARM;
8539: if (stepm<=12) stepsize=1;
8540: if(estepm < stepm){
8541: printf ("Problem %d lower than %d\n",estepm, stepm);
8542: }
1.270 brouard 8543: else{
8544: hstepm=estepm;
8545: }
8546: if(estepm >= stepm){ /* Yes every two year */
8547: stepsize=2;
8548: }
1.267 brouard 8549:
8550: hstepm=hstepm/stepm;
1.296 brouard 8551: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8552: /* fractional in yp1 *\/ */
8553: /* aintmean=yp; */
8554: /* yp2=modf((yp1*12),&yp); */
8555: /* mintmean=yp; */
8556: /* yp1=modf((yp2*30.5),&yp); */
8557: /* jintmean=yp; */
8558: /* if(jintmean==0) jintmean=1; */
8559: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8560:
8561: i1=pow(2,cptcoveff);
8562: if (cptcovn < 1){i1=1;}
8563:
1.296 brouard 8564: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8565: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8566:
8567: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8568:
8569: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8570: for(k=1; k<=i1;k++){
8571: if(i1 != 1 && TKresult[nres]!= k)
8572: continue;
8573: if(invalidvarcomb[k]){
8574: printf("\nCombination (%d) projection ignored because no cases \n",k);
8575: continue;
8576: }
1.268 brouard 8577: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8578: for(j=1;j<=cptcoveff;j++) {
8579: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8580: }
8581: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8582: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8583: }
8584: fprintf(ficresfb," yearbproj age");
8585: for(j=1; j<=nlstate+ndeath;j++){
8586: for(i=1; i<=nlstate;i++)
1.268 brouard 8587: fprintf(ficresfb," b%d%d",i,j);
8588: fprintf(ficresfb," b.%d",j);
1.267 brouard 8589: }
1.296 brouard 8590: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8591: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8592: fprintf(ficresfb,"\n");
1.296 brouard 8593: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8594: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8595: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8596: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8597: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8598: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8599: nhstepm = nhstepm/hstepm;
8600: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8601: oldm=oldms;savm=savms;
1.268 brouard 8602: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8603: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8604: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8605: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8606: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8607: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8608: for (h=0; h<=nhstepm; h++){
1.268 brouard 8609: if (h*hstepm/YEARM*stepm ==-yearp) {
8610: break;
8611: }
8612: }
8613: fprintf(ficresfb,"\n");
8614: for(j=1;j<=cptcoveff;j++)
8615: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8616: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8617: for(i=1; i<=nlstate+ndeath;i++) {
8618: ppij=0.;ppi=0.;
8619: for(j=1; j<=nlstate;j++) {
8620: /* if (mobilav==1) */
1.269 brouard 8621: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8622: ppi=ppi+prevacurrent[(int)agec][j][k];
8623: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8624: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8625: /* else { */
8626: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8627: /* } */
1.268 brouard 8628: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8629: } /* end j */
8630: if(ppi <0.99){
8631: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8632: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8633: }
8634: fprintf(ficresfb," %.3f", ppij);
8635: }/* end j */
1.267 brouard 8636: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8637: } /* end agec */
8638: } /* end yearp */
8639: } /* end k */
1.217 brouard 8640:
1.267 brouard 8641: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8642:
1.267 brouard 8643: fclose(ficresfb);
8644: printf("End of Computing Back forecasting \n");
8645: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8646:
1.267 brouard 8647: }
1.217 brouard 8648:
1.269 brouard 8649: /* Variance of prevalence limit: varprlim */
8650: 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){
1.288 brouard 8651: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8652:
8653: char fileresvpl[FILENAMELENGTH];
8654: FILE *ficresvpl;
8655: double **oldm, **savm;
8656: double **varpl; /* Variances of prevalence limits by age */
8657: int i1, k, nres, j ;
8658:
8659: strcpy(fileresvpl,"VPL_");
8660: strcat(fileresvpl,fileresu);
8661: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8662: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8663: exit(0);
8664: }
1.288 brouard 8665: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8666: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8667:
8668: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8669: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8670:
8671: i1=pow(2,cptcoveff);
8672: if (cptcovn < 1){i1=1;}
8673:
8674: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8675: for(k=1; k<=i1;k++){
8676: if(i1 != 1 && TKresult[nres]!= k)
8677: continue;
8678: fprintf(ficresvpl,"\n#****** ");
8679: printf("\n#****** ");
8680: fprintf(ficlog,"\n#****** ");
8681: for(j=1;j<=cptcoveff;j++) {
8682: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8683: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8684: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8685: }
8686: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8687: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8688: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8689: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8690: }
8691: fprintf(ficresvpl,"******\n");
8692: printf("******\n");
8693: fprintf(ficlog,"******\n");
8694:
8695: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8696: oldm=oldms;savm=savms;
8697: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8698: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8699: /*}*/
8700: }
8701:
8702: fclose(ficresvpl);
1.288 brouard 8703: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8704: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8705:
8706: }
8707: /* Variance of back prevalence: varbprlim */
8708: 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){
8709: /*------- Variance of back (stable) prevalence------*/
8710:
8711: char fileresvbl[FILENAMELENGTH];
8712: FILE *ficresvbl;
8713:
8714: double **oldm, **savm;
8715: double **varbpl; /* Variances of back prevalence limits by age */
8716: int i1, k, nres, j ;
8717:
8718: strcpy(fileresvbl,"VBL_");
8719: strcat(fileresvbl,fileresu);
8720: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8721: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8722: exit(0);
8723: }
8724: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8725: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8726:
8727:
8728: i1=pow(2,cptcoveff);
8729: if (cptcovn < 1){i1=1;}
8730:
8731: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8732: for(k=1; k<=i1;k++){
8733: if(i1 != 1 && TKresult[nres]!= k)
8734: continue;
8735: fprintf(ficresvbl,"\n#****** ");
8736: printf("\n#****** ");
8737: fprintf(ficlog,"\n#****** ");
8738: for(j=1;j<=cptcoveff;j++) {
8739: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8740: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8741: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8742: }
8743: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8744: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8745: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8746: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8747: }
8748: fprintf(ficresvbl,"******\n");
8749: printf("******\n");
8750: fprintf(ficlog,"******\n");
8751:
8752: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8753: oldm=oldms;savm=savms;
8754:
8755: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8756: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8757: /*}*/
8758: }
8759:
8760: fclose(ficresvbl);
8761: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8762: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8763:
8764: } /* End of varbprlim */
8765:
1.126 brouard 8766: /************** Forecasting *****not tested NB*************/
1.227 brouard 8767: /* 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 8768:
1.227 brouard 8769: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8770: /* int *popage; */
8771: /* double calagedatem, agelim, kk1, kk2; */
8772: /* double *popeffectif,*popcount; */
8773: /* double ***p3mat,***tabpop,***tabpopprev; */
8774: /* /\* double ***mobaverage; *\/ */
8775: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8776:
1.227 brouard 8777: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8778: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8779: /* agelim=AGESUP; */
8780: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8781:
1.227 brouard 8782: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8783:
8784:
1.227 brouard 8785: /* strcpy(filerespop,"POP_"); */
8786: /* strcat(filerespop,fileresu); */
8787: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8788: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8789: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8790: /* } */
8791: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8792: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8793:
1.227 brouard 8794: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8795:
1.227 brouard 8796: /* /\* if (mobilav!=0) { *\/ */
8797: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8798: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8799: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8800: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8801: /* /\* } *\/ */
8802: /* /\* } *\/ */
1.126 brouard 8803:
1.227 brouard 8804: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8805: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8806:
1.227 brouard 8807: /* agelim=AGESUP; */
1.126 brouard 8808:
1.227 brouard 8809: /* hstepm=1; */
8810: /* hstepm=hstepm/stepm; */
1.218 brouard 8811:
1.227 brouard 8812: /* if (popforecast==1) { */
8813: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8814: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8815: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8816: /* } */
8817: /* popage=ivector(0,AGESUP); */
8818: /* popeffectif=vector(0,AGESUP); */
8819: /* popcount=vector(0,AGESUP); */
1.126 brouard 8820:
1.227 brouard 8821: /* i=1; */
8822: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8823:
1.227 brouard 8824: /* imx=i; */
8825: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8826: /* } */
1.218 brouard 8827:
1.227 brouard 8828: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8829: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8830: /* k=k+1; */
8831: /* fprintf(ficrespop,"\n#******"); */
8832: /* for(j=1;j<=cptcoveff;j++) { */
8833: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8834: /* } */
8835: /* fprintf(ficrespop,"******\n"); */
8836: /* fprintf(ficrespop,"# Age"); */
8837: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8838: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8839:
1.227 brouard 8840: /* for (cpt=0; cpt<=0;cpt++) { */
8841: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8842:
1.227 brouard 8843: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8844: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8845: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8846:
1.227 brouard 8847: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8848: /* oldm=oldms;savm=savms; */
8849: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8850:
1.227 brouard 8851: /* for (h=0; h<=nhstepm; h++){ */
8852: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8853: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8854: /* } */
8855: /* for(j=1; j<=nlstate+ndeath;j++) { */
8856: /* kk1=0.;kk2=0; */
8857: /* for(i=1; i<=nlstate;i++) { */
8858: /* if (mobilav==1) */
8859: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8860: /* else { */
8861: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8862: /* } */
8863: /* } */
8864: /* if (h==(int)(calagedatem+12*cpt)){ */
8865: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8866: /* /\*fprintf(ficrespop," %.3f", kk1); */
8867: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8868: /* } */
8869: /* } */
8870: /* for(i=1; i<=nlstate;i++){ */
8871: /* kk1=0.; */
8872: /* for(j=1; j<=nlstate;j++){ */
8873: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8874: /* } */
8875: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8876: /* } */
1.218 brouard 8877:
1.227 brouard 8878: /* if (h==(int)(calagedatem+12*cpt)) */
8879: /* for(j=1; j<=nlstate;j++) */
8880: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8881: /* } */
8882: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8883: /* } */
8884: /* } */
1.218 brouard 8885:
1.227 brouard 8886: /* /\******\/ */
1.218 brouard 8887:
1.227 brouard 8888: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8889: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8890: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8891: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8892: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8893:
1.227 brouard 8894: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8895: /* oldm=oldms;savm=savms; */
8896: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8897: /* for (h=0; h<=nhstepm; h++){ */
8898: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8899: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8900: /* } */
8901: /* for(j=1; j<=nlstate+ndeath;j++) { */
8902: /* kk1=0.;kk2=0; */
8903: /* for(i=1; i<=nlstate;i++) { */
8904: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8905: /* } */
8906: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8907: /* } */
8908: /* } */
8909: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8910: /* } */
8911: /* } */
8912: /* } */
8913: /* } */
1.218 brouard 8914:
1.227 brouard 8915: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8916:
1.227 brouard 8917: /* if (popforecast==1) { */
8918: /* free_ivector(popage,0,AGESUP); */
8919: /* free_vector(popeffectif,0,AGESUP); */
8920: /* free_vector(popcount,0,AGESUP); */
8921: /* } */
8922: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8923: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8924: /* fclose(ficrespop); */
8925: /* } /\* End of popforecast *\/ */
1.218 brouard 8926:
1.126 brouard 8927: int fileappend(FILE *fichier, char *optionfich)
8928: {
8929: if((fichier=fopen(optionfich,"a"))==NULL) {
8930: printf("Problem with file: %s\n", optionfich);
8931: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8932: return (0);
8933: }
8934: fflush(fichier);
8935: return (1);
8936: }
8937:
8938:
8939: /**************** function prwizard **********************/
8940: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8941: {
8942:
8943: /* Wizard to print covariance matrix template */
8944:
1.164 brouard 8945: char ca[32], cb[32];
8946: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8947: int numlinepar;
8948:
8949: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8950: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8951: for(i=1; i <=nlstate; i++){
8952: jj=0;
8953: for(j=1; j <=nlstate+ndeath; j++){
8954: if(j==i) continue;
8955: jj++;
8956: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8957: printf("%1d%1d",i,j);
8958: fprintf(ficparo,"%1d%1d",i,j);
8959: for(k=1; k<=ncovmodel;k++){
8960: /* printf(" %lf",param[i][j][k]); */
8961: /* fprintf(ficparo," %lf",param[i][j][k]); */
8962: printf(" 0.");
8963: fprintf(ficparo," 0.");
8964: }
8965: printf("\n");
8966: fprintf(ficparo,"\n");
8967: }
8968: }
8969: printf("# Scales (for hessian or gradient estimation)\n");
8970: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8971: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8972: for(i=1; i <=nlstate; i++){
8973: jj=0;
8974: for(j=1; j <=nlstate+ndeath; j++){
8975: if(j==i) continue;
8976: jj++;
8977: fprintf(ficparo,"%1d%1d",i,j);
8978: printf("%1d%1d",i,j);
8979: fflush(stdout);
8980: for(k=1; k<=ncovmodel;k++){
8981: /* printf(" %le",delti3[i][j][k]); */
8982: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8983: printf(" 0.");
8984: fprintf(ficparo," 0.");
8985: }
8986: numlinepar++;
8987: printf("\n");
8988: fprintf(ficparo,"\n");
8989: }
8990: }
8991: printf("# Covariance matrix\n");
8992: /* # 121 Var(a12)\n\ */
8993: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8994: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8995: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8996: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8997: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8998: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8999: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9000: fflush(stdout);
9001: fprintf(ficparo,"# Covariance matrix\n");
9002: /* # 121 Var(a12)\n\ */
9003: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9004: /* # ...\n\ */
9005: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9006:
9007: for(itimes=1;itimes<=2;itimes++){
9008: jj=0;
9009: for(i=1; i <=nlstate; i++){
9010: for(j=1; j <=nlstate+ndeath; j++){
9011: if(j==i) continue;
9012: for(k=1; k<=ncovmodel;k++){
9013: jj++;
9014: ca[0]= k+'a'-1;ca[1]='\0';
9015: if(itimes==1){
9016: printf("#%1d%1d%d",i,j,k);
9017: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9018: }else{
9019: printf("%1d%1d%d",i,j,k);
9020: fprintf(ficparo,"%1d%1d%d",i,j,k);
9021: /* printf(" %.5le",matcov[i][j]); */
9022: }
9023: ll=0;
9024: for(li=1;li <=nlstate; li++){
9025: for(lj=1;lj <=nlstate+ndeath; lj++){
9026: if(lj==li) continue;
9027: for(lk=1;lk<=ncovmodel;lk++){
9028: ll++;
9029: if(ll<=jj){
9030: cb[0]= lk +'a'-1;cb[1]='\0';
9031: if(ll<jj){
9032: if(itimes==1){
9033: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9034: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9035: }else{
9036: printf(" 0.");
9037: fprintf(ficparo," 0.");
9038: }
9039: }else{
9040: if(itimes==1){
9041: printf(" Var(%s%1d%1d)",ca,i,j);
9042: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9043: }else{
9044: printf(" 0.");
9045: fprintf(ficparo," 0.");
9046: }
9047: }
9048: }
9049: } /* end lk */
9050: } /* end lj */
9051: } /* end li */
9052: printf("\n");
9053: fprintf(ficparo,"\n");
9054: numlinepar++;
9055: } /* end k*/
9056: } /*end j */
9057: } /* end i */
9058: } /* end itimes */
9059:
9060: } /* end of prwizard */
9061: /******************* Gompertz Likelihood ******************************/
9062: double gompertz(double x[])
9063: {
1.302 ! brouard 9064: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9065: int i,n=0; /* n is the size of the sample */
9066:
1.220 brouard 9067: for (i=1;i<=imx ; i++) {
1.126 brouard 9068: sump=sump+weight[i];
9069: /* sump=sump+1;*/
9070: num=num+1;
9071: }
1.302 ! brouard 9072: L=0.0;
! 9073: /* agegomp=AGEGOMP; */
1.126 brouard 9074: /* for (i=0; i<=imx; i++)
9075: 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]);*/
9076:
1.302 ! brouard 9077: for (i=1;i<=imx ; i++) {
! 9078: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
! 9079: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
! 9080: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
! 9081: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
! 9082: * +
! 9083: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
! 9084: */
! 9085: if (wav[i] > 1 || agedc[i] < AGESUP) {
! 9086: if (cens[i] == 1){
! 9087: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
! 9088: } else if (cens[i] == 0){
1.126 brouard 9089: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 ! brouard 9090: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
! 9091: } else
! 9092: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9093: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 ! brouard 9094: L=L+A*weight[i];
1.126 brouard 9095: /* 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]);*/
1.302 ! brouard 9096: }
! 9097: }
1.126 brouard 9098:
1.302 ! brouard 9099: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9100:
9101: return -2*L*num/sump;
9102: }
9103:
1.136 brouard 9104: #ifdef GSL
9105: /******************* Gompertz_f Likelihood ******************************/
9106: double gompertz_f(const gsl_vector *v, void *params)
9107: {
1.302 ! brouard 9108: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9109: double *x= (double *) v->data;
9110: int i,n=0; /* n is the size of the sample */
9111:
9112: for (i=0;i<=imx-1 ; i++) {
9113: sump=sump+weight[i];
9114: /* sump=sump+1;*/
9115: num=num+1;
9116: }
9117:
9118:
9119: /* for (i=0; i<=imx; i++)
9120: 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]);*/
9121: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9122: for (i=1;i<=imx ; i++)
9123: {
9124: if (cens[i] == 1 && wav[i]>1)
9125: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9126:
9127: if (cens[i] == 0 && wav[i]>1)
9128: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9129: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9130:
9131: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9132: if (wav[i] > 1 ) { /* ??? */
9133: LL=LL+A*weight[i];
9134: /* 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]);*/
9135: }
9136: }
9137:
9138: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9139: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9140:
9141: return -2*LL*num/sump;
9142: }
9143: #endif
9144:
1.126 brouard 9145: /******************* Printing html file ***********/
1.201 brouard 9146: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9147: int lastpass, int stepm, int weightopt, char model[],\
9148: int imx, double p[],double **matcov,double agemortsup){
9149: int i,k;
9150:
9151: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9152: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9153: for (i=1;i<=2;i++)
9154: 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 9155: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9156: fprintf(fichtm,"</ul>");
9157:
9158: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9159:
9160: 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>");
9161:
9162: for (k=agegomp;k<(agemortsup-2);k++)
9163: 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]);
9164:
9165:
9166: fflush(fichtm);
9167: }
9168:
9169: /******************* Gnuplot file **************/
1.201 brouard 9170: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9171:
9172: char dirfileres[132],optfileres[132];
1.164 brouard 9173:
1.126 brouard 9174: int ng;
9175:
9176:
9177: /*#ifdef windows */
9178: fprintf(ficgp,"cd \"%s\" \n",pathc);
9179: /*#endif */
9180:
9181:
9182: strcpy(dirfileres,optionfilefiname);
9183: strcpy(optfileres,"vpl");
1.199 brouard 9184: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9185: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9186: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9187: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9188: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9189:
9190: }
9191:
1.136 brouard 9192: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9193: {
1.126 brouard 9194:
1.136 brouard 9195: /*-------- data file ----------*/
9196: FILE *fic;
9197: char dummy[]=" ";
1.240 brouard 9198: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9199: int lstra;
1.136 brouard 9200: int linei, month, year,iout;
1.302 ! brouard 9201: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9202: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9203: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9204: char *stratrunc;
1.223 brouard 9205:
1.240 brouard 9206: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9207: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9208:
1.240 brouard 9209: for(v=1; v <=ncovcol;v++){
9210: DummyV[v]=0;
9211: FixedV[v]=0;
9212: }
9213: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9214: DummyV[v]=1;
9215: FixedV[v]=0;
9216: }
9217: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9218: DummyV[v]=0;
9219: FixedV[v]=1;
9220: }
9221: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9222: DummyV[v]=1;
9223: FixedV[v]=1;
9224: }
9225: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9226: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9227: 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]);
9228: }
1.126 brouard 9229:
1.136 brouard 9230: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9231: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9232: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9233: }
1.126 brouard 9234:
1.302 ! brouard 9235: /* Is it a BOM UTF-8 Windows file? */
! 9236: /* First data line */
! 9237: linei=0;
! 9238: while(fgets(line, MAXLINE, fic)) {
! 9239: noffset=0;
! 9240: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
! 9241: {
! 9242: noffset=noffset+3;
! 9243: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
! 9244: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
! 9245: fflush(ficlog); return 1;
! 9246: }
! 9247: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
! 9248: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
! 9249: {
! 9250: noffset=noffset+2;
! 9251: printf("# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
! 9252: fprintf(ficlog,"# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
! 9253: fflush(ficlog); return 1;
! 9254: }
! 9255: else if( line[0] == 0 && line[1] == 0)
! 9256: {
! 9257: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
! 9258: noffset=noffset+4;
! 9259: printf("# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
! 9260: fprintf(ficlog,"# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
! 9261: fflush(ficlog); return 1;
! 9262: }
! 9263: } else{
! 9264: ;/*printf(" Not a BOM file\n");*/
! 9265: }
! 9266: /* If line starts with a # it is a comment */
! 9267: if (line[noffset] == '#') {
! 9268: linei=linei+1;
! 9269: break;
! 9270: }else{
! 9271: break;
! 9272: }
! 9273: }
! 9274: fclose(fic);
! 9275: if((fic=fopen(datafile,"r"))==NULL) {
! 9276: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
! 9277: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
! 9278: }
! 9279: /* Not a Bom file */
! 9280:
1.136 brouard 9281: i=1;
9282: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9283: linei=linei+1;
9284: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9285: if(line[j] == '\t')
9286: line[j] = ' ';
9287: }
9288: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9289: ;
9290: };
9291: line[j+1]=0; /* Trims blanks at end of line */
9292: if(line[0]=='#'){
9293: fprintf(ficlog,"Comment line\n%s\n",line);
9294: printf("Comment line\n%s\n",line);
9295: continue;
9296: }
9297: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9298: strcpy(line, linetmp);
1.223 brouard 9299:
9300: /* Loops on waves */
9301: for (j=maxwav;j>=1;j--){
9302: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9303: cutv(stra, strb, line, ' ');
9304: if(strb[0]=='.') { /* Missing value */
9305: lval=-1;
9306: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9307: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9308: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9309: 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);
9310: 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);
9311: return 1;
9312: }
9313: }else{
9314: errno=0;
9315: /* what_kind_of_number(strb); */
9316: dval=strtod(strb,&endptr);
9317: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9318: /* if(strb != endptr && *endptr == '\0') */
9319: /* dval=dlval; */
9320: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9321: if( strb[0]=='\0' || (*endptr != '\0')){
9322: 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);
9323: 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);
9324: return 1;
9325: }
9326: cotqvar[j][iv][i]=dval;
9327: cotvar[j][ntv+iv][i]=dval;
9328: }
9329: strcpy(line,stra);
1.223 brouard 9330: }/* end loop ntqv */
1.225 brouard 9331:
1.223 brouard 9332: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9333: cutv(stra, strb, line, ' ');
9334: if(strb[0]=='.') { /* Missing value */
9335: lval=-1;
9336: }else{
9337: errno=0;
9338: lval=strtol(strb,&endptr,10);
9339: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9340: if( strb[0]=='\0' || (*endptr != '\0')){
9341: 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);
9342: 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);
9343: return 1;
9344: }
9345: }
9346: if(lval <-1 || lval >1){
9347: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9348: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9349: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9350: For example, for multinomial values like 1, 2 and 3,\n \
9351: build V1=0 V2=0 for the reference value (1),\n \
9352: V1=1 V2=0 for (2) \n \
1.223 brouard 9353: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9354: output of IMaCh is often meaningless.\n \
1.223 brouard 9355: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9356: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9357: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9358: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9359: For example, for multinomial values like 1, 2 and 3,\n \
9360: build V1=0 V2=0 for the reference value (1),\n \
9361: V1=1 V2=0 for (2) \n \
1.223 brouard 9362: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9363: output of IMaCh is often meaningless.\n \
1.223 brouard 9364: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9365: return 1;
9366: }
9367: cotvar[j][iv][i]=(double)(lval);
9368: strcpy(line,stra);
1.223 brouard 9369: }/* end loop ntv */
1.225 brouard 9370:
1.223 brouard 9371: /* Statuses at wave */
1.137 brouard 9372: cutv(stra, strb, line, ' ');
1.223 brouard 9373: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9374: lval=-1;
1.136 brouard 9375: }else{
1.238 brouard 9376: errno=0;
9377: lval=strtol(strb,&endptr,10);
9378: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9379: if( strb[0]=='\0' || (*endptr != '\0')){
9380: 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);
9381: 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);
9382: return 1;
9383: }
1.136 brouard 9384: }
1.225 brouard 9385:
1.136 brouard 9386: s[j][i]=lval;
1.225 brouard 9387:
1.223 brouard 9388: /* Date of Interview */
1.136 brouard 9389: strcpy(line,stra);
9390: cutv(stra, strb,line,' ');
1.169 brouard 9391: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9392: }
1.169 brouard 9393: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9394: month=99;
9395: year=9999;
1.136 brouard 9396: }else{
1.225 brouard 9397: 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);
9398: 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);
9399: return 1;
1.136 brouard 9400: }
9401: anint[j][i]= (double) year;
1.302 ! brouard 9402: mint[j][i]= (double)month;
! 9403: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
! 9404: /* printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
! 9405: /* fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
! 9406: /* } */
1.136 brouard 9407: strcpy(line,stra);
1.223 brouard 9408: } /* End loop on waves */
1.225 brouard 9409:
1.223 brouard 9410: /* Date of death */
1.136 brouard 9411: cutv(stra, strb,line,' ');
1.169 brouard 9412: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9413: }
1.169 brouard 9414: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9415: month=99;
9416: year=9999;
9417: }else{
1.141 brouard 9418: 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 9419: 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);
9420: return 1;
1.136 brouard 9421: }
9422: andc[i]=(double) year;
9423: moisdc[i]=(double) month;
9424: strcpy(line,stra);
9425:
1.223 brouard 9426: /* Date of birth */
1.136 brouard 9427: cutv(stra, strb,line,' ');
1.169 brouard 9428: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9429: }
1.169 brouard 9430: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9431: month=99;
9432: year=9999;
9433: }else{
1.141 brouard 9434: 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);
9435: 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 9436: return 1;
1.136 brouard 9437: }
9438: if (year==9999) {
1.141 brouard 9439: 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);
9440: 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 9441: return 1;
9442:
1.136 brouard 9443: }
9444: annais[i]=(double)(year);
1.302 ! brouard 9445: moisnais[i]=(double)(month);
! 9446: for (j=1;j<=maxwav;j++){
! 9447: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
! 9448: printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j,(int)moisnais[i],(int)annais[i]);
! 9449: fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j, (int)moisnais[i],(int)annais[i]);
! 9450: }
! 9451: }
! 9452:
1.136 brouard 9453: strcpy(line,stra);
1.225 brouard 9454:
1.223 brouard 9455: /* Sample weight */
1.136 brouard 9456: cutv(stra, strb,line,' ');
9457: errno=0;
9458: dval=strtod(strb,&endptr);
9459: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9460: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9461: 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 9462: fflush(ficlog);
9463: return 1;
9464: }
9465: weight[i]=dval;
9466: strcpy(line,stra);
1.225 brouard 9467:
1.223 brouard 9468: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9469: cutv(stra, strb, line, ' ');
9470: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9471: lval=-1;
1.223 brouard 9472: }else{
1.225 brouard 9473: errno=0;
9474: /* what_kind_of_number(strb); */
9475: dval=strtod(strb,&endptr);
9476: /* if(strb != endptr && *endptr == '\0') */
9477: /* dval=dlval; */
9478: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9479: if( strb[0]=='\0' || (*endptr != '\0')){
9480: 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);
9481: 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);
9482: return 1;
9483: }
9484: coqvar[iv][i]=dval;
1.226 brouard 9485: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9486: }
9487: strcpy(line,stra);
9488: }/* end loop nqv */
1.136 brouard 9489:
1.223 brouard 9490: /* Covariate values */
1.136 brouard 9491: for (j=ncovcol;j>=1;j--){
9492: cutv(stra, strb,line,' ');
1.223 brouard 9493: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9494: lval=-1;
1.136 brouard 9495: }else{
1.225 brouard 9496: errno=0;
9497: lval=strtol(strb,&endptr,10);
9498: if( strb[0]=='\0' || (*endptr != '\0')){
9499: 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);
9500: 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);
9501: return 1;
9502: }
1.136 brouard 9503: }
9504: if(lval <-1 || lval >1){
1.225 brouard 9505: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9506: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9507: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9508: For example, for multinomial values like 1, 2 and 3,\n \
9509: build V1=0 V2=0 for the reference value (1),\n \
9510: V1=1 V2=0 for (2) \n \
1.136 brouard 9511: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9512: output of IMaCh is often meaningless.\n \
1.136 brouard 9513: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9514: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9515: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9516: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9517: For example, for multinomial values like 1, 2 and 3,\n \
9518: build V1=0 V2=0 for the reference value (1),\n \
9519: V1=1 V2=0 for (2) \n \
1.136 brouard 9520: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9521: output of IMaCh is often meaningless.\n \
1.136 brouard 9522: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9523: return 1;
1.136 brouard 9524: }
9525: covar[j][i]=(double)(lval);
9526: strcpy(line,stra);
9527: }
9528: lstra=strlen(stra);
1.225 brouard 9529:
1.136 brouard 9530: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9531: stratrunc = &(stra[lstra-9]);
9532: num[i]=atol(stratrunc);
9533: }
9534: else
9535: num[i]=atol(stra);
9536: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9537: 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;}*/
9538:
9539: i=i+1;
9540: } /* End loop reading data */
1.225 brouard 9541:
1.136 brouard 9542: *imax=i-1; /* Number of individuals */
9543: fclose(fic);
1.225 brouard 9544:
1.136 brouard 9545: return (0);
1.164 brouard 9546: /* endread: */
1.225 brouard 9547: printf("Exiting readdata: ");
9548: fclose(fic);
9549: return (1);
1.223 brouard 9550: }
1.126 brouard 9551:
1.234 brouard 9552: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9553: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9554: while (*p2 == ' ')
1.234 brouard 9555: p2++;
9556: /* while ((*p1++ = *p2++) !=0) */
9557: /* ; */
9558: /* do */
9559: /* while (*p2 == ' ') */
9560: /* p2++; */
9561: /* while (*p1++ == *p2++); */
9562: *stri=p2;
1.145 brouard 9563: }
9564:
1.235 brouard 9565: int decoderesult ( char resultline[], int nres)
1.230 brouard 9566: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9567: {
1.235 brouard 9568: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9569: char resultsav[MAXLINE];
1.234 brouard 9570: int resultmodel[MAXLINE];
9571: int modelresult[MAXLINE];
1.230 brouard 9572: char stra[80], strb[80], strc[80], strd[80],stre[80];
9573:
1.234 brouard 9574: removefirstspace(&resultline);
1.233 brouard 9575: printf("decoderesult:%s\n",resultline);
1.230 brouard 9576:
9577: if (strstr(resultline,"v") !=0){
9578: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9579: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9580: return 1;
9581: }
9582: trimbb(resultsav, resultline);
9583: if (strlen(resultsav) >1){
9584: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9585: }
1.253 brouard 9586: if(j == 0){ /* Resultline but no = */
9587: TKresult[nres]=0; /* Combination for the nresult and the model */
9588: return (0);
9589: }
9590:
1.234 brouard 9591: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9592: 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);
9593: 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);
9594: }
9595: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9596: if(nbocc(resultsav,'=') >1){
9597: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9598: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9599: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9600: }else
9601: cutl(strc,strd,resultsav,'=');
1.230 brouard 9602: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9603:
1.230 brouard 9604: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9605: Tvarsel[k]=atoi(strc);
9606: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9607: /* cptcovsel++; */
9608: if (nbocc(stra,'=') >0)
9609: strcpy(resultsav,stra); /* and analyzes it */
9610: }
1.235 brouard 9611: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9612: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9613: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9614: match=0;
1.236 brouard 9615: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9616: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9617: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9618: match=1;
9619: break;
9620: }
9621: }
9622: if(match == 0){
9623: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9624: }
9625: }
9626: }
1.235 brouard 9627: /* Checking for missing or useless values in comparison of current model needs */
9628: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9629: match=0;
1.235 brouard 9630: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9631: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9632: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9633: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9634: ++match;
9635: }
9636: }
9637: }
9638: if(match == 0){
9639: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9640: }else if(match > 1){
9641: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9642: }
9643: }
1.235 brouard 9644:
1.234 brouard 9645: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9646: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9647: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9648: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9649: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9650: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9651: /* 1 0 0 0 */
9652: /* 2 1 0 0 */
9653: /* 3 0 1 0 */
9654: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9655: /* 5 0 0 1 */
9656: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9657: /* 7 0 1 1 */
9658: /* 8 1 1 1 */
1.237 brouard 9659: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9660: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9661: /* V5*age V5 known which value for nres? */
9662: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9663: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9664: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9665: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9666: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9667: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9668: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9669: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9670: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9671: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9672: k4++;;
9673: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9674: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9675: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9676: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9677: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9678: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9679: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9680: k4q++;;
9681: }
9682: }
1.234 brouard 9683:
1.235 brouard 9684: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9685: return (0);
9686: }
1.235 brouard 9687:
1.230 brouard 9688: int decodemodel( char model[], int lastobs)
9689: /**< This routine decodes the model and returns:
1.224 brouard 9690: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9691: * - nagesqr = 1 if age*age in the model, otherwise 0.
9692: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9693: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9694: * - cptcovage number of covariates with age*products =2
9695: * - cptcovs number of simple covariates
9696: * - 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
9697: * which is a new column after the 9 (ncovcol) variables.
9698: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9699: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9700: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9701: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9702: */
1.136 brouard 9703: {
1.238 brouard 9704: int i, j, k, ks, v;
1.227 brouard 9705: int j1, k1, k2, k3, k4;
1.136 brouard 9706: char modelsav[80];
1.145 brouard 9707: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9708: char *strpt;
1.136 brouard 9709:
1.145 brouard 9710: /*removespace(model);*/
1.136 brouard 9711: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9712: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9713: if (strstr(model,"AGE") !=0){
1.192 brouard 9714: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9715: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9716: return 1;
9717: }
1.141 brouard 9718: if (strstr(model,"v") !=0){
9719: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9720: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9721: return 1;
9722: }
1.187 brouard 9723: strcpy(modelsav,model);
9724: if ((strpt=strstr(model,"age*age")) !=0){
9725: printf(" strpt=%s, model=%s\n",strpt, model);
9726: if(strpt != model){
1.234 brouard 9727: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9728: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9729: corresponding column of parameters.\n",model);
1.234 brouard 9730: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9731: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9732: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9733: return 1;
1.225 brouard 9734: }
1.187 brouard 9735: nagesqr=1;
9736: if (strstr(model,"+age*age") !=0)
1.234 brouard 9737: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9738: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9739: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9740: else
1.234 brouard 9741: substrchaine(modelsav, model, "age*age");
1.187 brouard 9742: }else
9743: nagesqr=0;
9744: if (strlen(modelsav) >1){
9745: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9746: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9747: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9748: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9749: * cst, age and age*age
9750: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9751: /* including age products which are counted in cptcovage.
9752: * but the covariates which are products must be treated
9753: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9754: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9755: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9756:
9757:
1.187 brouard 9758: /* Design
9759: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9760: * < ncovcol=8 >
9761: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9762: * k= 1 2 3 4 5 6 7 8
9763: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9764: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9765: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9766: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9767: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9768: * Tage[++cptcovage]=k
9769: * if products, new covar are created after ncovcol with k1
9770: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9771: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9772: * 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
9773: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9774: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9775: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9776: * < ncovcol=8 >
9777: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9778: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9779: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9780: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9781: * p Tprod[1]@2={ 6, 5}
9782: *p Tvard[1][1]@4= {7, 8, 5, 6}
9783: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9784: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9785: *How to reorganize?
9786: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9787: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9788: * {2, 1, 4, 8, 5, 6, 3, 7}
9789: * Struct []
9790: */
1.225 brouard 9791:
1.187 brouard 9792: /* This loop fills the array Tvar from the string 'model'.*/
9793: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9794: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9795: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9796: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9797: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9798: /* k=1 Tvar[1]=2 (from V2) */
9799: /* k=5 Tvar[5] */
9800: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9801: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9802: /* } */
1.198 brouard 9803: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9804: /*
9805: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9806: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9807: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9808: }
1.187 brouard 9809: cptcovage=0;
9810: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9811: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9812: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9813: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9814: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9815: /*scanf("%d",i);*/
9816: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9817: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9818: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9819: /* covar is not filled and then is empty */
9820: cptcovprod--;
9821: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9822: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9823: Typevar[k]=1; /* 1 for age product */
9824: cptcovage++; /* Sums the number of covariates which include age as a product */
9825: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9826: /*printf("stre=%s ", stre);*/
9827: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9828: cptcovprod--;
9829: cutl(stre,strb,strc,'V');
9830: Tvar[k]=atoi(stre);
9831: Typevar[k]=1; /* 1 for age product */
9832: cptcovage++;
9833: Tage[cptcovage]=k;
9834: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9835: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9836: cptcovn++;
9837: cptcovprodnoage++;k1++;
9838: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9839: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9840: because this model-covariate is a construction we invent a new column
9841: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9842: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9843: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9844: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9845: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9846: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9847: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9848: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9849: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9850: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9851: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9852: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9853: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9854: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9855: for (i=1; i<=lastobs;i++){
9856: /* Computes the new covariate which is a product of
9857: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9858: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9859: }
9860: } /* End age is not in the model */
9861: } /* End if model includes a product */
9862: else { /* no more sum */
9863: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9864: /* scanf("%d",i);*/
9865: cutl(strd,strc,strb,'V');
9866: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9867: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9868: Tvar[k]=atoi(strd);
9869: Typevar[k]=0; /* 0 for simple covariates */
9870: }
9871: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9872: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9873: scanf("%d",i);*/
1.187 brouard 9874: } /* end of loop + on total covariates */
9875: } /* end if strlen(modelsave == 0) age*age might exist */
9876: } /* end if strlen(model == 0) */
1.136 brouard 9877:
9878: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9879: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9880:
1.136 brouard 9881: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9882: printf("cptcovprod=%d ", cptcovprod);
9883: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9884: scanf("%d ",i);*/
9885:
9886:
1.230 brouard 9887: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9888: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9889: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9890: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9891: k = 1 2 3 4 5 6 7 8 9
9892: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9893: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9894: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9895: Dummy[k] 1 0 0 0 3 1 1 2 3
9896: Tmodelind[combination of covar]=k;
1.225 brouard 9897: */
9898: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9899: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9900: /* 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 9901: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9902: printf("Model=%s\n\
9903: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9904: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9905: 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);
9906: fprintf(ficlog,"Model=%s\n\
9907: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9908: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9909: 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);
1.285 brouard 9910: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9911: 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 */
9912: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9913: Fixed[k]= 0;
9914: Dummy[k]= 0;
1.225 brouard 9915: ncoveff++;
1.232 brouard 9916: ncovf++;
1.234 brouard 9917: nsd++;
9918: modell[k].maintype= FTYPE;
9919: TvarsD[nsd]=Tvar[k];
9920: TvarsDind[nsd]=k;
9921: TvarF[ncovf]=Tvar[k];
9922: TvarFind[ncovf]=k;
9923: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9924: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9925: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9926: Fixed[k]= 0;
9927: Dummy[k]= 0;
9928: ncoveff++;
9929: ncovf++;
9930: modell[k].maintype= FTYPE;
9931: TvarF[ncovf]=Tvar[k];
9932: TvarFind[ncovf]=k;
1.230 brouard 9933: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9934: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9935: }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 9936: Fixed[k]= 0;
9937: Dummy[k]= 1;
1.230 brouard 9938: nqfveff++;
1.234 brouard 9939: modell[k].maintype= FTYPE;
9940: modell[k].subtype= FQ;
9941: nsq++;
9942: TvarsQ[nsq]=Tvar[k];
9943: TvarsQind[nsq]=k;
1.232 brouard 9944: ncovf++;
1.234 brouard 9945: TvarF[ncovf]=Tvar[k];
9946: TvarFind[ncovf]=k;
1.231 brouard 9947: 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 9948: 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 9949: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9950: Fixed[k]= 1;
9951: Dummy[k]= 0;
1.225 brouard 9952: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9953: modell[k].maintype= VTYPE;
9954: modell[k].subtype= VD;
9955: nsd++;
9956: TvarsD[nsd]=Tvar[k];
9957: TvarsDind[nsd]=k;
9958: ncovv++; /* Only simple time varying variables */
9959: TvarV[ncovv]=Tvar[k];
1.242 brouard 9960: 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 9961: 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 */
9962: 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 9963: 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);
9964: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9965: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9966: Fixed[k]= 1;
9967: Dummy[k]= 1;
9968: nqtveff++;
9969: modell[k].maintype= VTYPE;
9970: modell[k].subtype= VQ;
9971: ncovv++; /* Only simple time varying variables */
9972: nsq++;
9973: TvarsQ[nsq]=Tvar[k];
9974: TvarsQind[nsq]=k;
9975: TvarV[ncovv]=Tvar[k];
1.242 brouard 9976: 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 9977: 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 */
9978: 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 9979: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9980: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9981: 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 9982: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9983: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9984: ncova++;
9985: TvarA[ncova]=Tvar[k];
9986: TvarAind[ncova]=k;
1.231 brouard 9987: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9988: Fixed[k]= 2;
9989: Dummy[k]= 2;
9990: modell[k].maintype= ATYPE;
9991: modell[k].subtype= APFD;
9992: /* ncoveff++; */
1.227 brouard 9993: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9994: Fixed[k]= 2;
9995: Dummy[k]= 3;
9996: modell[k].maintype= ATYPE;
9997: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9998: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9999: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10000: Fixed[k]= 3;
10001: Dummy[k]= 2;
10002: modell[k].maintype= ATYPE;
10003: modell[k].subtype= APVD; /* Product age * varying dummy */
10004: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10005: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10006: Fixed[k]= 3;
10007: Dummy[k]= 3;
10008: modell[k].maintype= ATYPE;
10009: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10010: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10011: }
10012: }else if (Typevar[k] == 2) { /* product without age */
10013: k1=Tposprod[k];
10014: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10015: if(Tvard[k1][2] <=ncovcol){
10016: Fixed[k]= 1;
10017: Dummy[k]= 0;
10018: modell[k].maintype= FTYPE;
10019: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10020: ncovf++; /* Fixed variables without age */
10021: TvarF[ncovf]=Tvar[k];
10022: TvarFind[ncovf]=k;
10023: }else if(Tvard[k1][2] <=ncovcol+nqv){
10024: Fixed[k]= 0; /* or 2 ?*/
10025: Dummy[k]= 1;
10026: modell[k].maintype= FTYPE;
10027: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10028: ncovf++; /* Varying variables without age */
10029: TvarF[ncovf]=Tvar[k];
10030: TvarFind[ncovf]=k;
10031: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10032: Fixed[k]= 1;
10033: Dummy[k]= 0;
10034: modell[k].maintype= VTYPE;
10035: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10036: ncovv++; /* Varying variables without age */
10037: TvarV[ncovv]=Tvar[k];
10038: TvarVind[ncovv]=k;
10039: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10040: Fixed[k]= 1;
10041: Dummy[k]= 1;
10042: modell[k].maintype= VTYPE;
10043: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10044: ncovv++; /* Varying variables without age */
10045: TvarV[ncovv]=Tvar[k];
10046: TvarVind[ncovv]=k;
10047: }
1.227 brouard 10048: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10049: if(Tvard[k1][2] <=ncovcol){
10050: Fixed[k]= 0; /* or 2 ?*/
10051: Dummy[k]= 1;
10052: modell[k].maintype= FTYPE;
10053: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10054: ncovf++; /* Fixed variables without age */
10055: TvarF[ncovf]=Tvar[k];
10056: TvarFind[ncovf]=k;
10057: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10058: Fixed[k]= 1;
10059: Dummy[k]= 1;
10060: modell[k].maintype= VTYPE;
10061: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10062: ncovv++; /* Varying variables without age */
10063: TvarV[ncovv]=Tvar[k];
10064: TvarVind[ncovv]=k;
10065: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10066: Fixed[k]= 1;
10067: Dummy[k]= 1;
10068: modell[k].maintype= VTYPE;
10069: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10070: ncovv++; /* Varying variables without age */
10071: TvarV[ncovv]=Tvar[k];
10072: TvarVind[ncovv]=k;
10073: ncovv++; /* Varying variables without age */
10074: TvarV[ncovv]=Tvar[k];
10075: TvarVind[ncovv]=k;
10076: }
1.227 brouard 10077: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10078: if(Tvard[k1][2] <=ncovcol){
10079: Fixed[k]= 1;
10080: Dummy[k]= 1;
10081: modell[k].maintype= VTYPE;
10082: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10083: ncovv++; /* Varying variables without age */
10084: TvarV[ncovv]=Tvar[k];
10085: TvarVind[ncovv]=k;
10086: }else if(Tvard[k1][2] <=ncovcol+nqv){
10087: Fixed[k]= 1;
10088: Dummy[k]= 1;
10089: modell[k].maintype= VTYPE;
10090: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10091: ncovv++; /* Varying variables without age */
10092: TvarV[ncovv]=Tvar[k];
10093: TvarVind[ncovv]=k;
10094: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10095: Fixed[k]= 1;
10096: Dummy[k]= 0;
10097: modell[k].maintype= VTYPE;
10098: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10099: ncovv++; /* Varying variables without age */
10100: TvarV[ncovv]=Tvar[k];
10101: TvarVind[ncovv]=k;
10102: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10103: Fixed[k]= 1;
10104: Dummy[k]= 1;
10105: modell[k].maintype= VTYPE;
10106: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10107: ncovv++; /* Varying variables without age */
10108: TvarV[ncovv]=Tvar[k];
10109: TvarVind[ncovv]=k;
10110: }
1.227 brouard 10111: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10112: if(Tvard[k1][2] <=ncovcol){
10113: Fixed[k]= 1;
10114: Dummy[k]= 1;
10115: modell[k].maintype= VTYPE;
10116: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10117: ncovv++; /* Varying variables without age */
10118: TvarV[ncovv]=Tvar[k];
10119: TvarVind[ncovv]=k;
10120: }else if(Tvard[k1][2] <=ncovcol+nqv){
10121: Fixed[k]= 1;
10122: Dummy[k]= 1;
10123: modell[k].maintype= VTYPE;
10124: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10125: ncovv++; /* Varying variables without age */
10126: TvarV[ncovv]=Tvar[k];
10127: TvarVind[ncovv]=k;
10128: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10129: Fixed[k]= 1;
10130: Dummy[k]= 1;
10131: modell[k].maintype= VTYPE;
10132: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10133: ncovv++; /* Varying variables without age */
10134: TvarV[ncovv]=Tvar[k];
10135: TvarVind[ncovv]=k;
10136: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10137: Fixed[k]= 1;
10138: Dummy[k]= 1;
10139: modell[k].maintype= VTYPE;
10140: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10141: ncovv++; /* Varying variables without age */
10142: TvarV[ncovv]=Tvar[k];
10143: TvarVind[ncovv]=k;
10144: }
1.227 brouard 10145: }else{
1.240 brouard 10146: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10147: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10148: } /*end k1*/
1.225 brouard 10149: }else{
1.226 brouard 10150: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10151: 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 10152: }
1.227 brouard 10153: 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 10154: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10155: 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]);
10156: }
10157: /* Searching for doublons in the model */
10158: for(k1=1; k1<= cptcovt;k1++){
10159: for(k2=1; k2 <k1;k2++){
1.285 brouard 10160: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10161: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10162: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10163: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10164: 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[k1],Dummy[k1]);
10165: 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[k1],Dummy[k1]); fflush(ficlog);
1.234 brouard 10166: return(1);
10167: }
10168: }else if (Typevar[k1] ==2){
10169: k3=Tposprod[k1];
10170: k4=Tposprod[k2];
10171: 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])) ){
10172: 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]]);
10173: 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);
10174: return(1);
10175: }
10176: }
1.227 brouard 10177: }
10178: }
1.225 brouard 10179: }
10180: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10181: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10182: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10183: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10184: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10185: /*endread:*/
1.225 brouard 10186: printf("Exiting decodemodel: ");
10187: return (1);
1.136 brouard 10188: }
10189:
1.169 brouard 10190: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10191: {/* Check ages at death */
1.136 brouard 10192: int i, m;
1.218 brouard 10193: int firstone=0;
10194:
1.136 brouard 10195: for (i=1; i<=imx; i++) {
10196: for(m=2; (m<= maxwav); m++) {
10197: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10198: anint[m][i]=9999;
1.216 brouard 10199: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10200: s[m][i]=-1;
1.136 brouard 10201: }
10202: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10203: *nberr = *nberr + 1;
1.218 brouard 10204: if(firstone == 0){
10205: firstone=1;
1.260 brouard 10206: 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 10207: }
1.262 brouard 10208: 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 10209: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10210: }
10211: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10212: (*nberr)++;
1.259 brouard 10213: 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 10214: 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 10215: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10216: }
10217: }
10218: }
10219:
10220: for (i=1; i<=imx; i++) {
10221: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10222: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10223: 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 10224: if (s[m][i] >= nlstate+1) {
1.169 brouard 10225: if(agedc[i]>0){
10226: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10227: agev[m][i]=agedc[i];
1.214 brouard 10228: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10229: }else {
1.136 brouard 10230: if ((int)andc[i]!=9999){
10231: nbwarn++;
10232: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10233: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10234: agev[m][i]=-1;
10235: }
10236: }
1.169 brouard 10237: } /* agedc > 0 */
1.214 brouard 10238: } /* end if */
1.136 brouard 10239: else if(s[m][i] !=9){ /* Standard case, age in fractional
10240: years but with the precision of a month */
10241: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10242: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10243: agev[m][i]=1;
10244: else if(agev[m][i] < *agemin){
10245: *agemin=agev[m][i];
10246: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10247: }
10248: else if(agev[m][i] >*agemax){
10249: *agemax=agev[m][i];
1.156 brouard 10250: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10251: }
10252: /*agev[m][i]=anint[m][i]-annais[i];*/
10253: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10254: } /* en if 9*/
1.136 brouard 10255: else { /* =9 */
1.214 brouard 10256: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10257: agev[m][i]=1;
10258: s[m][i]=-1;
10259: }
10260: }
1.214 brouard 10261: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10262: agev[m][i]=1;
1.214 brouard 10263: else{
10264: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10265: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10266: agev[m][i]=0;
10267: }
10268: } /* End for lastpass */
10269: }
1.136 brouard 10270:
10271: for (i=1; i<=imx; i++) {
10272: for(m=firstpass; (m<=lastpass); m++){
10273: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10274: (*nberr)++;
1.136 brouard 10275: 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);
10276: 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);
10277: return 1;
10278: }
10279: }
10280: }
10281:
10282: /*for (i=1; i<=imx; i++){
10283: for (m=firstpass; (m<lastpass); m++){
10284: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10285: }
10286:
10287: }*/
10288:
10289:
1.139 brouard 10290: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10291: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10292:
10293: return (0);
1.164 brouard 10294: /* endread:*/
1.136 brouard 10295: printf("Exiting calandcheckages: ");
10296: return (1);
10297: }
10298:
1.172 brouard 10299: #if defined(_MSC_VER)
10300: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10301: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10302: //#include "stdafx.h"
10303: //#include <stdio.h>
10304: //#include <tchar.h>
10305: //#include <windows.h>
10306: //#include <iostream>
10307: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10308:
10309: LPFN_ISWOW64PROCESS fnIsWow64Process;
10310:
10311: BOOL IsWow64()
10312: {
10313: BOOL bIsWow64 = FALSE;
10314:
10315: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10316: // (HANDLE, PBOOL);
10317:
10318: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10319:
10320: HMODULE module = GetModuleHandle(_T("kernel32"));
10321: const char funcName[] = "IsWow64Process";
10322: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10323: GetProcAddress(module, funcName);
10324:
10325: if (NULL != fnIsWow64Process)
10326: {
10327: if (!fnIsWow64Process(GetCurrentProcess(),
10328: &bIsWow64))
10329: //throw std::exception("Unknown error");
10330: printf("Unknown error\n");
10331: }
10332: return bIsWow64 != FALSE;
10333: }
10334: #endif
1.177 brouard 10335:
1.191 brouard 10336: void syscompilerinfo(int logged)
1.292 brouard 10337: {
10338: #include <stdint.h>
10339:
10340: /* #include "syscompilerinfo.h"*/
1.185 brouard 10341: /* command line Intel compiler 32bit windows, XP compatible:*/
10342: /* /GS /W3 /Gy
10343: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10344: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10345: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10346: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10347: */
10348: /* 64 bits */
1.185 brouard 10349: /*
10350: /GS /W3 /Gy
10351: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10352: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10353: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10354: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10355: /* Optimization are useless and O3 is slower than O2 */
10356: /*
10357: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10358: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10359: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10360: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10361: */
1.186 brouard 10362: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10363: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10364: /PDB:"visual studio
10365: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10366: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10367: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10368: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10369: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10370: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10371: uiAccess='false'"
10372: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10373: /NOLOGO /TLBID:1
10374: */
1.292 brouard 10375:
10376:
1.177 brouard 10377: #if defined __INTEL_COMPILER
1.178 brouard 10378: #if defined(__GNUC__)
10379: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10380: #endif
1.177 brouard 10381: #elif defined(__GNUC__)
1.179 brouard 10382: #ifndef __APPLE__
1.174 brouard 10383: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10384: #endif
1.177 brouard 10385: struct utsname sysInfo;
1.178 brouard 10386: int cross = CROSS;
10387: if (cross){
10388: printf("Cross-");
1.191 brouard 10389: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10390: }
1.174 brouard 10391: #endif
10392:
1.191 brouard 10393: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10394: #if defined(__clang__)
1.191 brouard 10395: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10396: #endif
10397: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10398: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10399: #endif
10400: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10401: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10402: #endif
10403: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10404: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10405: #endif
10406: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10407: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10408: #endif
10409: #if defined(_MSC_VER)
1.191 brouard 10410: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10411: #endif
10412: #if defined(__PGI)
1.191 brouard 10413: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10414: #endif
10415: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10416: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10417: #endif
1.191 brouard 10418: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10419:
1.167 brouard 10420: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10421: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10422: // Windows (x64 and x86)
1.191 brouard 10423: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10424: #elif __unix__ // all unices, not all compilers
10425: // Unix
1.191 brouard 10426: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10427: #elif __linux__
10428: // linux
1.191 brouard 10429: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10430: #elif __APPLE__
1.174 brouard 10431: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10432: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10433: #endif
10434:
10435: /* __MINGW32__ */
10436: /* __CYGWIN__ */
10437: /* __MINGW64__ */
10438: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10439: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10440: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10441: /* _WIN64 // Defined for applications for Win64. */
10442: /* _M_X64 // Defined for compilations that target x64 processors. */
10443: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10444:
1.167 brouard 10445: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10446: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10447: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10448: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10449: #else
1.191 brouard 10450: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10451: #endif
10452:
1.169 brouard 10453: #if defined(__GNUC__)
10454: # if defined(__GNUC_PATCHLEVEL__)
10455: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10456: + __GNUC_MINOR__ * 100 \
10457: + __GNUC_PATCHLEVEL__)
10458: # else
10459: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10460: + __GNUC_MINOR__ * 100)
10461: # endif
1.174 brouard 10462: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10463: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10464:
10465: if (uname(&sysInfo) != -1) {
10466: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10467: 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 10468: }
10469: else
10470: perror("uname() error");
1.179 brouard 10471: //#ifndef __INTEL_COMPILER
10472: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10473: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10474: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10475: #endif
1.169 brouard 10476: #endif
1.172 brouard 10477:
1.286 brouard 10478: // void main ()
1.172 brouard 10479: // {
1.169 brouard 10480: #if defined(_MSC_VER)
1.174 brouard 10481: if (IsWow64()){
1.191 brouard 10482: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10483: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10484: }
10485: else{
1.191 brouard 10486: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10487: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10488: }
1.172 brouard 10489: // printf("\nPress Enter to continue...");
10490: // getchar();
10491: // }
10492:
1.169 brouard 10493: #endif
10494:
1.167 brouard 10495:
1.219 brouard 10496: }
1.136 brouard 10497:
1.219 brouard 10498: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10499: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10500: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10501: /* double ftolpl = 1.e-10; */
1.180 brouard 10502: double age, agebase, agelim;
1.203 brouard 10503: double tot;
1.180 brouard 10504:
1.202 brouard 10505: strcpy(filerespl,"PL_");
10506: strcat(filerespl,fileresu);
10507: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10508: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10509: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10510: }
1.288 brouard 10511: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10512: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10513: pstamp(ficrespl);
1.288 brouard 10514: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10515: fprintf(ficrespl,"#Age ");
10516: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10517: fprintf(ficrespl,"\n");
1.180 brouard 10518:
1.219 brouard 10519: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10520:
1.219 brouard 10521: agebase=ageminpar;
10522: agelim=agemaxpar;
1.180 brouard 10523:
1.227 brouard 10524: /* i1=pow(2,ncoveff); */
1.234 brouard 10525: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10526: if (cptcovn < 1){i1=1;}
1.180 brouard 10527:
1.238 brouard 10528: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10529: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10530: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10531: continue;
1.235 brouard 10532:
1.238 brouard 10533: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10534: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10535: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10536: /* k=k+1; */
10537: /* to clean */
10538: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10539: fprintf(ficrespl,"#******");
10540: printf("#******");
10541: fprintf(ficlog,"#******");
10542: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10543: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10544: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10545: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10546: }
10547: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10548: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10549: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10550: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10551: }
10552: fprintf(ficrespl,"******\n");
10553: printf("******\n");
10554: fprintf(ficlog,"******\n");
10555: if(invalidvarcomb[k]){
10556: printf("\nCombination (%d) ignored because no case \n",k);
10557: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10558: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10559: continue;
10560: }
1.219 brouard 10561:
1.238 brouard 10562: fprintf(ficrespl,"#Age ");
10563: for(j=1;j<=cptcoveff;j++) {
10564: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10565: }
10566: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10567: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10568:
1.238 brouard 10569: for (age=agebase; age<=agelim; age++){
10570: /* for (age=agebase; age<=agebase; age++){ */
10571: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10572: fprintf(ficrespl,"%.0f ",age );
10573: for(j=1;j<=cptcoveff;j++)
10574: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10575: tot=0.;
10576: for(i=1; i<=nlstate;i++){
10577: tot += prlim[i][i];
10578: fprintf(ficrespl," %.5f", prlim[i][i]);
10579: }
10580: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10581: } /* Age */
10582: /* was end of cptcod */
10583: } /* cptcov */
10584: } /* nres */
1.219 brouard 10585: return 0;
1.180 brouard 10586: }
10587:
1.218 brouard 10588: 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){
1.288 brouard 10589: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10590:
10591: /* Computes the back prevalence limit for any combination of covariate values
10592: * at any age between ageminpar and agemaxpar
10593: */
1.235 brouard 10594: int i, j, k, i1, nres=0 ;
1.217 brouard 10595: /* double ftolpl = 1.e-10; */
10596: double age, agebase, agelim;
10597: double tot;
1.218 brouard 10598: /* double ***mobaverage; */
10599: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10600:
10601: strcpy(fileresplb,"PLB_");
10602: strcat(fileresplb,fileresu);
10603: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10604: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10605: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10606: }
1.288 brouard 10607: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10608: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10609: pstamp(ficresplb);
1.288 brouard 10610: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10611: fprintf(ficresplb,"#Age ");
10612: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10613: fprintf(ficresplb,"\n");
10614:
1.218 brouard 10615:
10616: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10617:
10618: agebase=ageminpar;
10619: agelim=agemaxpar;
10620:
10621:
1.227 brouard 10622: i1=pow(2,cptcoveff);
1.218 brouard 10623: if (cptcovn < 1){i1=1;}
1.227 brouard 10624:
1.238 brouard 10625: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10626: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10627: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10628: continue;
10629: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10630: fprintf(ficresplb,"#******");
10631: printf("#******");
10632: fprintf(ficlog,"#******");
10633: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10634: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10635: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10636: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10637: }
10638: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10639: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10640: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10641: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10642: }
10643: fprintf(ficresplb,"******\n");
10644: printf("******\n");
10645: fprintf(ficlog,"******\n");
10646: if(invalidvarcomb[k]){
10647: printf("\nCombination (%d) ignored because no cases \n",k);
10648: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10649: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10650: continue;
10651: }
1.218 brouard 10652:
1.238 brouard 10653: fprintf(ficresplb,"#Age ");
10654: for(j=1;j<=cptcoveff;j++) {
10655: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10656: }
10657: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10658: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10659:
10660:
1.238 brouard 10661: for (age=agebase; age<=agelim; age++){
10662: /* for (age=agebase; age<=agebase; age++){ */
10663: if(mobilavproj > 0){
10664: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10665: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10666: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10667: }else if (mobilavproj == 0){
10668: 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);
10669: 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);
10670: exit(1);
10671: }else{
10672: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10673: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10674: /* printf("TOTOT\n"); */
10675: /* exit(1); */
1.238 brouard 10676: }
10677: fprintf(ficresplb,"%.0f ",age );
10678: for(j=1;j<=cptcoveff;j++)
10679: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10680: tot=0.;
10681: for(i=1; i<=nlstate;i++){
10682: tot += bprlim[i][i];
10683: fprintf(ficresplb," %.5f", bprlim[i][i]);
10684: }
10685: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10686: } /* Age */
10687: /* was end of cptcod */
1.255 brouard 10688: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10689: } /* end of any combination */
10690: } /* end of nres */
1.218 brouard 10691: /* hBijx(p, bage, fage); */
10692: /* fclose(ficrespijb); */
10693:
10694: return 0;
1.217 brouard 10695: }
1.218 brouard 10696:
1.180 brouard 10697: int hPijx(double *p, int bage, int fage){
10698: /*------------- h Pij x at various ages ------------*/
10699:
10700: int stepsize;
10701: int agelim;
10702: int hstepm;
10703: int nhstepm;
1.235 brouard 10704: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10705:
10706: double agedeb;
10707: double ***p3mat;
10708:
1.201 brouard 10709: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10710: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10711: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10712: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10713: }
10714: printf("Computing pij: result on file '%s' \n", filerespij);
10715: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10716:
10717: stepsize=(int) (stepm+YEARM-1)/YEARM;
10718: /*if (stepm<=24) stepsize=2;*/
10719:
10720: agelim=AGESUP;
10721: hstepm=stepsize*YEARM; /* Every year of age */
10722: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10723:
1.180 brouard 10724: /* hstepm=1; aff par mois*/
10725: pstamp(ficrespij);
10726: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10727: i1= pow(2,cptcoveff);
1.218 brouard 10728: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10729: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10730: /* k=k+1; */
1.235 brouard 10731: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10732: for(k=1; k<=i1;k++){
1.253 brouard 10733: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10734: continue;
1.183 brouard 10735: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10736: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10737: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10738: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10739: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10740: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10741: }
1.183 brouard 10742: fprintf(ficrespij,"******\n");
10743:
10744: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10745: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10746: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10747:
10748: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10749:
1.183 brouard 10750: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10751: oldm=oldms;savm=savms;
1.235 brouard 10752: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10753: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10754: for(i=1; i<=nlstate;i++)
10755: for(j=1; j<=nlstate+ndeath;j++)
10756: fprintf(ficrespij," %1d-%1d",i,j);
10757: fprintf(ficrespij,"\n");
10758: for (h=0; h<=nhstepm; h++){
10759: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10760: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10761: for(i=1; i<=nlstate;i++)
10762: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10763: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10764: fprintf(ficrespij,"\n");
10765: }
1.183 brouard 10766: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10767: fprintf(ficrespij,"\n");
10768: }
1.180 brouard 10769: /*}*/
10770: }
1.218 brouard 10771: return 0;
1.180 brouard 10772: }
1.218 brouard 10773:
10774: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10775: /*------------- h Bij x at various ages ------------*/
10776:
10777: int stepsize;
1.218 brouard 10778: /* int agelim; */
10779: int ageminl;
1.217 brouard 10780: int hstepm;
10781: int nhstepm;
1.238 brouard 10782: int h, i, i1, j, k, nres;
1.218 brouard 10783:
1.217 brouard 10784: double agedeb;
10785: double ***p3mat;
1.218 brouard 10786:
10787: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10788: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10789: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10790: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10791: }
10792: printf("Computing pij back: result on file '%s' \n", filerespijb);
10793: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10794:
10795: stepsize=(int) (stepm+YEARM-1)/YEARM;
10796: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10797:
1.218 brouard 10798: /* agelim=AGESUP; */
1.289 brouard 10799: ageminl=AGEINF; /* was 30 */
1.218 brouard 10800: hstepm=stepsize*YEARM; /* Every year of age */
10801: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10802:
10803: /* hstepm=1; aff par mois*/
10804: pstamp(ficrespijb);
1.255 brouard 10805: 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 10806: i1= pow(2,cptcoveff);
1.218 brouard 10807: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10808: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10809: /* k=k+1; */
1.238 brouard 10810: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10811: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10812: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10813: continue;
10814: fprintf(ficrespijb,"\n#****** ");
10815: for(j=1;j<=cptcoveff;j++)
10816: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10817: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10818: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10819: }
10820: fprintf(ficrespijb,"******\n");
1.264 brouard 10821: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10822: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10823: continue;
10824: }
10825:
10826: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10827: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10828: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10829: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm+0.1)-1; /* Typically 20 years = 20*12/6=40 or 55*12/24=27.5-1.1=>27 */
10830: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10831:
10832: /* nhstepm=nhstepm*YEARM; aff par mois*/
10833:
1.266 brouard 10834: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10835: /* and memory limitations if stepm is small */
10836:
1.238 brouard 10837: /* oldm=oldms;savm=savms; */
10838: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10839: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10840: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10841: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10842: for(i=1; i<=nlstate;i++)
10843: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10844: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10845: fprintf(ficrespijb,"\n");
1.238 brouard 10846: for (h=0; h<=nhstepm; h++){
10847: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10848: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10849: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10850: for(i=1; i<=nlstate;i++)
10851: for(j=1; j<=nlstate+ndeath;j++)
10852: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10853: fprintf(ficrespijb,"\n");
10854: }
10855: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10856: fprintf(ficrespijb,"\n");
10857: } /* end age deb */
10858: } /* end combination */
10859: } /* end nres */
1.218 brouard 10860: return 0;
10861: } /* hBijx */
1.217 brouard 10862:
1.180 brouard 10863:
1.136 brouard 10864: /***********************************************/
10865: /**************** Main Program *****************/
10866: /***********************************************/
10867:
10868: int main(int argc, char *argv[])
10869: {
10870: #ifdef GSL
10871: const gsl_multimin_fminimizer_type *T;
10872: size_t iteri = 0, it;
10873: int rval = GSL_CONTINUE;
10874: int status = GSL_SUCCESS;
10875: double ssval;
10876: #endif
10877: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10878: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10879: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10880: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10881: int jj, ll, li, lj, lk;
1.136 brouard 10882: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10883: int num_filled;
1.136 brouard 10884: int itimes;
10885: int NDIM=2;
10886: int vpopbased=0;
1.235 brouard 10887: int nres=0;
1.258 brouard 10888: int endishere=0;
1.277 brouard 10889: int noffset=0;
1.274 brouard 10890: int ncurrv=0; /* Temporary variable */
10891:
1.164 brouard 10892: char ca[32], cb[32];
1.136 brouard 10893: /* FILE *fichtm; *//* Html File */
10894: /* FILE *ficgp;*/ /*Gnuplot File */
10895: struct stat info;
1.191 brouard 10896: double agedeb=0.;
1.194 brouard 10897:
10898: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10899: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10900:
1.165 brouard 10901: double fret;
1.191 brouard 10902: double dum=0.; /* Dummy variable */
1.136 brouard 10903: double ***p3mat;
1.218 brouard 10904: /* double ***mobaverage; */
1.164 brouard 10905:
10906: char line[MAXLINE];
1.197 brouard 10907: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10908:
1.234 brouard 10909: char modeltemp[MAXLINE];
1.230 brouard 10910: char resultline[MAXLINE];
10911:
1.136 brouard 10912: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10913: char *tok, *val; /* pathtot */
1.290 brouard 10914: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10915: int c, h , cpt, c2;
1.191 brouard 10916: int jl=0;
10917: int i1, j1, jk, stepsize=0;
1.194 brouard 10918: int count=0;
10919:
1.164 brouard 10920: int *tab;
1.136 brouard 10921: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10922: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10923: /* double anprojf, mprojf, jprojf; */
10924: /* double jintmean,mintmean,aintmean; */
10925: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10926: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10927: double yrfproj= 10.0; /* Number of years of forward projections */
10928: double yrbproj= 10.0; /* Number of years of backward projections */
10929: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 10930: int mobilav=0,popforecast=0;
1.191 brouard 10931: int hstepm=0, nhstepm=0;
1.136 brouard 10932: int agemortsup;
10933: float sumlpop=0.;
10934: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10935: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10936:
1.191 brouard 10937: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10938: double ftolpl=FTOL;
10939: double **prlim;
1.217 brouard 10940: double **bprlim;
1.136 brouard 10941: double ***param; /* Matrix of parameters */
1.251 brouard 10942: double ***paramstart; /* Matrix of starting parameter values */
10943: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10944: double **matcov; /* Matrix of covariance */
1.203 brouard 10945: double **hess; /* Hessian matrix */
1.136 brouard 10946: double ***delti3; /* Scale */
10947: double *delti; /* Scale */
10948: double ***eij, ***vareij;
10949: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10950:
1.136 brouard 10951: double *epj, vepp;
1.164 brouard 10952:
1.273 brouard 10953: double dateprev1, dateprev2;
1.296 brouard 10954: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
10955: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
10956:
1.217 brouard 10957:
1.136 brouard 10958: double **ximort;
1.145 brouard 10959: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10960: int *dcwave;
10961:
1.164 brouard 10962: char z[1]="c";
1.136 brouard 10963:
10964: /*char *strt;*/
10965: char strtend[80];
1.126 brouard 10966:
1.164 brouard 10967:
1.126 brouard 10968: /* setlocale (LC_ALL, ""); */
10969: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10970: /* textdomain (PACKAGE); */
10971: /* setlocale (LC_CTYPE, ""); */
10972: /* setlocale (LC_MESSAGES, ""); */
10973:
10974: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10975: rstart_time = time(NULL);
10976: /* (void) gettimeofday(&start_time,&tzp);*/
10977: start_time = *localtime(&rstart_time);
1.126 brouard 10978: curr_time=start_time;
1.157 brouard 10979: /*tml = *localtime(&start_time.tm_sec);*/
10980: /* strcpy(strstart,asctime(&tml)); */
10981: strcpy(strstart,asctime(&start_time));
1.126 brouard 10982:
10983: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10984: /* tp.tm_sec = tp.tm_sec +86400; */
10985: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10986: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10987: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10988: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10989: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10990: /* strt=asctime(&tmg); */
10991: /* printf("Time(after) =%s",strstart); */
10992: /* (void) time (&time_value);
10993: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10994: * tm = *localtime(&time_value);
10995: * strstart=asctime(&tm);
10996: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10997: */
10998:
10999: nberr=0; /* Number of errors and warnings */
11000: nbwarn=0;
1.184 brouard 11001: #ifdef WIN32
11002: _getcwd(pathcd, size);
11003: #else
1.126 brouard 11004: getcwd(pathcd, size);
1.184 brouard 11005: #endif
1.191 brouard 11006: syscompilerinfo(0);
1.196 brouard 11007: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11008: if(argc <=1){
11009: printf("\nEnter the parameter file name: ");
1.205 brouard 11010: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11011: printf("ERROR Empty parameter file name\n");
11012: goto end;
11013: }
1.126 brouard 11014: i=strlen(pathr);
11015: if(pathr[i-1]=='\n')
11016: pathr[i-1]='\0';
1.156 brouard 11017: i=strlen(pathr);
1.205 brouard 11018: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11019: pathr[i-1]='\0';
1.205 brouard 11020: }
11021: i=strlen(pathr);
11022: if( i==0 ){
11023: printf("ERROR Empty parameter file name\n");
11024: goto end;
11025: }
11026: for (tok = pathr; tok != NULL; ){
1.126 brouard 11027: printf("Pathr |%s|\n",pathr);
11028: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11029: printf("val= |%s| pathr=%s\n",val,pathr);
11030: strcpy (pathtot, val);
11031: if(pathr[0] == '\0') break; /* Dirty */
11032: }
11033: }
1.281 brouard 11034: else if (argc<=2){
11035: strcpy(pathtot,argv[1]);
11036: }
1.126 brouard 11037: else{
11038: strcpy(pathtot,argv[1]);
1.281 brouard 11039: strcpy(z,argv[2]);
11040: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11041: }
11042: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11043: /*cygwin_split_path(pathtot,path,optionfile);
11044: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11045: /* cutv(path,optionfile,pathtot,'\\');*/
11046:
11047: /* Split argv[0], imach program to get pathimach */
11048: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11049: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11050: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11051: /* strcpy(pathimach,argv[0]); */
11052: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11053: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11054: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11055: #ifdef WIN32
11056: _chdir(path); /* Can be a relative path */
11057: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11058: #else
1.126 brouard 11059: chdir(path); /* Can be a relative path */
1.184 brouard 11060: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11061: #endif
11062: printf("Current directory %s!\n",pathcd);
1.126 brouard 11063: strcpy(command,"mkdir ");
11064: strcat(command,optionfilefiname);
11065: if((outcmd=system(command)) != 0){
1.169 brouard 11066: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11067: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11068: /* fclose(ficlog); */
11069: /* exit(1); */
11070: }
11071: /* if((imk=mkdir(optionfilefiname))<0){ */
11072: /* perror("mkdir"); */
11073: /* } */
11074:
11075: /*-------- arguments in the command line --------*/
11076:
1.186 brouard 11077: /* Main Log file */
1.126 brouard 11078: strcat(filelog, optionfilefiname);
11079: strcat(filelog,".log"); /* */
11080: if((ficlog=fopen(filelog,"w"))==NULL) {
11081: printf("Problem with logfile %s\n",filelog);
11082: goto end;
11083: }
11084: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11085: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11086: fprintf(ficlog,"\nEnter the parameter file name: \n");
11087: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11088: path=%s \n\
11089: optionfile=%s\n\
11090: optionfilext=%s\n\
1.156 brouard 11091: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11092:
1.197 brouard 11093: syscompilerinfo(1);
1.167 brouard 11094:
1.126 brouard 11095: printf("Local time (at start):%s",strstart);
11096: fprintf(ficlog,"Local time (at start): %s",strstart);
11097: fflush(ficlog);
11098: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11099: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11100:
11101: /* */
11102: strcpy(fileres,"r");
11103: strcat(fileres, optionfilefiname);
1.201 brouard 11104: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11105: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11106: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11107:
1.186 brouard 11108: /* Main ---------arguments file --------*/
1.126 brouard 11109:
11110: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11111: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11112: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11113: fflush(ficlog);
1.149 brouard 11114: /* goto end; */
11115: exit(70);
1.126 brouard 11116: }
11117:
11118: strcpy(filereso,"o");
1.201 brouard 11119: strcat(filereso,fileresu);
1.126 brouard 11120: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11121: printf("Problem with Output resultfile: %s\n", filereso);
11122: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11123: fflush(ficlog);
11124: goto end;
11125: }
1.278 brouard 11126: /*-------- Rewriting parameter file ----------*/
11127: strcpy(rfileres,"r"); /* "Rparameterfile */
11128: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11129: strcat(rfileres,"."); /* */
11130: strcat(rfileres,optionfilext); /* Other files have txt extension */
11131: if((ficres =fopen(rfileres,"w"))==NULL) {
11132: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11133: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11134: fflush(ficlog);
11135: goto end;
11136: }
11137: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11138:
1.278 brouard 11139:
1.126 brouard 11140: /* Reads comments: lines beginning with '#' */
11141: numlinepar=0;
1.277 brouard 11142: /* Is it a BOM UTF-8 Windows file? */
11143: /* First parameter line */
1.197 brouard 11144: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11145: noffset=0;
11146: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11147: {
11148: noffset=noffset+3;
11149: printf("# File is an UTF8 Bom.\n"); // 0xBF
11150: }
1.302 ! brouard 11151: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
! 11152: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11153: {
11154: noffset=noffset+2;
11155: printf("# File is an UTF16BE BOM file\n");
11156: }
11157: else if( line[0] == 0 && line[1] == 0)
11158: {
11159: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11160: noffset=noffset+4;
11161: printf("# File is an UTF16BE BOM file\n");
11162: }
11163: } else{
11164: ;/*printf(" Not a BOM file\n");*/
11165: }
11166:
1.197 brouard 11167: /* If line starts with a # it is a comment */
1.277 brouard 11168: if (line[noffset] == '#') {
1.197 brouard 11169: numlinepar++;
11170: fputs(line,stdout);
11171: fputs(line,ficparo);
1.278 brouard 11172: fputs(line,ficres);
1.197 brouard 11173: fputs(line,ficlog);
11174: continue;
11175: }else
11176: break;
11177: }
11178: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11179: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11180: if (num_filled != 5) {
11181: printf("Should be 5 parameters\n");
1.283 brouard 11182: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11183: }
1.126 brouard 11184: numlinepar++;
1.197 brouard 11185: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11186: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11187: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11188: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11189: }
11190: /* Second parameter line */
11191: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11192: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11193: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11194: if (line[0] == '#') {
11195: numlinepar++;
1.283 brouard 11196: printf("%s",line);
11197: fprintf(ficres,"%s",line);
11198: fprintf(ficparo,"%s",line);
11199: fprintf(ficlog,"%s",line);
1.197 brouard 11200: continue;
11201: }else
11202: break;
11203: }
1.223 brouard 11204: 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", \
11205: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11206: if (num_filled != 11) {
11207: 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 11208: printf("but line=%s\n",line);
1.283 brouard 11209: fprintf(ficlog,"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");
11210: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11211: }
1.286 brouard 11212: if( lastpass > maxwav){
11213: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11214: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11215: fflush(ficlog);
11216: goto end;
11217: }
11218: 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);
1.283 brouard 11219: fprintf(ficparo,"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);
1.286 brouard 11220: fprintf(ficres,"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, 0, weightopt);
1.283 brouard 11221: fprintf(ficlog,"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 11222: }
1.203 brouard 11223: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11224: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11225: /* Third parameter line */
11226: while(fgets(line, MAXLINE, ficpar)) {
11227: /* If line starts with a # it is a comment */
11228: if (line[0] == '#') {
11229: numlinepar++;
1.283 brouard 11230: printf("%s",line);
11231: fprintf(ficres,"%s",line);
11232: fprintf(ficparo,"%s",line);
11233: fprintf(ficlog,"%s",line);
1.197 brouard 11234: continue;
11235: }else
11236: break;
11237: }
1.201 brouard 11238: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11239: if (num_filled != 1){
1.302 ! brouard 11240: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
! 11241: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11242: model[0]='\0';
11243: goto end;
11244: }
11245: else{
11246: if (model[0]=='+'){
11247: for(i=1; i<=strlen(model);i++)
11248: modeltemp[i-1]=model[i];
1.201 brouard 11249: strcpy(model,modeltemp);
1.197 brouard 11250: }
11251: }
1.199 brouard 11252: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11253: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11254: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11255: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11256: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11257: }
11258: /* 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); */
11259: /* numlinepar=numlinepar+3; /\* In general *\/ */
11260: /* 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); */
1.283 brouard 11261: /* 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); */
11262: /* 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 11263: fflush(ficlog);
1.190 brouard 11264: /* if(model[0]=='#'|| model[0]== '\0'){ */
11265: if(model[0]=='#'){
1.279 brouard 11266: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11267: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11268: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11269: if(mle != -1){
1.279 brouard 11270: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter vectors and subdiagonal covariance matrix.\n");
1.187 brouard 11271: exit(1);
11272: }
11273: }
1.126 brouard 11274: while((c=getc(ficpar))=='#' && c!= EOF){
11275: ungetc(c,ficpar);
11276: fgets(line, MAXLINE, ficpar);
11277: numlinepar++;
1.195 brouard 11278: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11279: z[0]=line[1];
11280: }
11281: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11282: fputs(line, stdout);
11283: //puts(line);
1.126 brouard 11284: fputs(line,ficparo);
11285: fputs(line,ficlog);
11286: }
11287: ungetc(c,ficpar);
11288:
11289:
1.290 brouard 11290: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11291: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11292: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11293: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11294: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11295: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11296: v1+v2*age+v2*v3 makes cptcovn = 3
11297: */
11298: if (strlen(model)>1)
1.187 brouard 11299: 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 11300: else
1.187 brouard 11301: ncovmodel=2; /* Constant and age */
1.133 brouard 11302: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11303: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11304: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11305: 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);
11306: 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);
11307: fflush(stdout);
11308: fclose (ficlog);
11309: goto end;
11310: }
1.126 brouard 11311: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11312: delti=delti3[1][1];
11313: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11314: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11315: /* We could also provide initial parameters values giving by simple logistic regression
11316: * only one way, that is without matrix product. We will have nlstate maximizations */
11317: /* for(i=1;i<nlstate;i++){ */
11318: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11319: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11320: /* } */
1.126 brouard 11321: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11322: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11323: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11324: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11325: fclose (ficparo);
11326: fclose (ficlog);
11327: goto end;
11328: exit(0);
1.220 brouard 11329: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11330: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11331: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11332: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11333: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11334: matcov=matrix(1,npar,1,npar);
1.203 brouard 11335: hess=matrix(1,npar,1,npar);
1.220 brouard 11336: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11337: /* Read guessed parameters */
1.126 brouard 11338: /* Reads comments: lines beginning with '#' */
11339: while((c=getc(ficpar))=='#' && c!= EOF){
11340: ungetc(c,ficpar);
11341: fgets(line, MAXLINE, ficpar);
11342: numlinepar++;
1.141 brouard 11343: fputs(line,stdout);
1.126 brouard 11344: fputs(line,ficparo);
11345: fputs(line,ficlog);
11346: }
11347: ungetc(c,ficpar);
11348:
11349: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11350: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11351: for(i=1; i <=nlstate; i++){
1.234 brouard 11352: j=0;
1.126 brouard 11353: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11354: if(jj==i) continue;
11355: j++;
1.292 brouard 11356: while((c=getc(ficpar))=='#' && c!= EOF){
11357: ungetc(c,ficpar);
11358: fgets(line, MAXLINE, ficpar);
11359: numlinepar++;
11360: fputs(line,stdout);
11361: fputs(line,ficparo);
11362: fputs(line,ficlog);
11363: }
11364: ungetc(c,ficpar);
1.234 brouard 11365: fscanf(ficpar,"%1d%1d",&i1,&j1);
11366: if ((i1 != i) || (j1 != jj)){
11367: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11368: It might be a problem of design; if ncovcol and the model are correct\n \
11369: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11370: exit(1);
11371: }
11372: fprintf(ficparo,"%1d%1d",i1,j1);
11373: if(mle==1)
11374: printf("%1d%1d",i,jj);
11375: fprintf(ficlog,"%1d%1d",i,jj);
11376: for(k=1; k<=ncovmodel;k++){
11377: fscanf(ficpar," %lf",¶m[i][j][k]);
11378: if(mle==1){
11379: printf(" %lf",param[i][j][k]);
11380: fprintf(ficlog," %lf",param[i][j][k]);
11381: }
11382: else
11383: fprintf(ficlog," %lf",param[i][j][k]);
11384: fprintf(ficparo," %lf",param[i][j][k]);
11385: }
11386: fscanf(ficpar,"\n");
11387: numlinepar++;
11388: if(mle==1)
11389: printf("\n");
11390: fprintf(ficlog,"\n");
11391: fprintf(ficparo,"\n");
1.126 brouard 11392: }
11393: }
11394: fflush(ficlog);
1.234 brouard 11395:
1.251 brouard 11396: /* Reads parameters values */
1.126 brouard 11397: p=param[1][1];
1.251 brouard 11398: pstart=paramstart[1][1];
1.126 brouard 11399:
11400: /* Reads comments: lines beginning with '#' */
11401: while((c=getc(ficpar))=='#' && c!= EOF){
11402: ungetc(c,ficpar);
11403: fgets(line, MAXLINE, ficpar);
11404: numlinepar++;
1.141 brouard 11405: fputs(line,stdout);
1.126 brouard 11406: fputs(line,ficparo);
11407: fputs(line,ficlog);
11408: }
11409: ungetc(c,ficpar);
11410:
11411: for(i=1; i <=nlstate; i++){
11412: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11413: fscanf(ficpar,"%1d%1d",&i1,&j1);
11414: if ( (i1-i) * (j1-j) != 0){
11415: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11416: exit(1);
11417: }
11418: printf("%1d%1d",i,j);
11419: fprintf(ficparo,"%1d%1d",i1,j1);
11420: fprintf(ficlog,"%1d%1d",i1,j1);
11421: for(k=1; k<=ncovmodel;k++){
11422: fscanf(ficpar,"%le",&delti3[i][j][k]);
11423: printf(" %le",delti3[i][j][k]);
11424: fprintf(ficparo," %le",delti3[i][j][k]);
11425: fprintf(ficlog," %le",delti3[i][j][k]);
11426: }
11427: fscanf(ficpar,"\n");
11428: numlinepar++;
11429: printf("\n");
11430: fprintf(ficparo,"\n");
11431: fprintf(ficlog,"\n");
1.126 brouard 11432: }
11433: }
11434: fflush(ficlog);
1.234 brouard 11435:
1.145 brouard 11436: /* Reads covariance matrix */
1.126 brouard 11437: delti=delti3[1][1];
1.220 brouard 11438:
11439:
1.126 brouard 11440: /* 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 11441:
1.126 brouard 11442: /* Reads comments: lines beginning with '#' */
11443: while((c=getc(ficpar))=='#' && c!= EOF){
11444: ungetc(c,ficpar);
11445: fgets(line, MAXLINE, ficpar);
11446: numlinepar++;
1.141 brouard 11447: fputs(line,stdout);
1.126 brouard 11448: fputs(line,ficparo);
11449: fputs(line,ficlog);
11450: }
11451: ungetc(c,ficpar);
1.220 brouard 11452:
1.126 brouard 11453: matcov=matrix(1,npar,1,npar);
1.203 brouard 11454: hess=matrix(1,npar,1,npar);
1.131 brouard 11455: for(i=1; i <=npar; i++)
11456: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11457:
1.194 brouard 11458: /* Scans npar lines */
1.126 brouard 11459: for(i=1; i <=npar; i++){
1.226 brouard 11460: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11461: if(count != 3){
1.226 brouard 11462: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11463: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11464: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11465: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11466: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11467: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11468: exit(1);
1.220 brouard 11469: }else{
1.226 brouard 11470: if(mle==1)
11471: printf("%1d%1d%d",i1,j1,jk);
11472: }
11473: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11474: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11475: for(j=1; j <=i; j++){
1.226 brouard 11476: fscanf(ficpar," %le",&matcov[i][j]);
11477: if(mle==1){
11478: printf(" %.5le",matcov[i][j]);
11479: }
11480: fprintf(ficlog," %.5le",matcov[i][j]);
11481: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11482: }
11483: fscanf(ficpar,"\n");
11484: numlinepar++;
11485: if(mle==1)
1.220 brouard 11486: printf("\n");
1.126 brouard 11487: fprintf(ficlog,"\n");
11488: fprintf(ficparo,"\n");
11489: }
1.194 brouard 11490: /* End of read covariance matrix npar lines */
1.126 brouard 11491: for(i=1; i <=npar; i++)
11492: for(j=i+1;j<=npar;j++)
1.226 brouard 11493: matcov[i][j]=matcov[j][i];
1.126 brouard 11494:
11495: if(mle==1)
11496: printf("\n");
11497: fprintf(ficlog,"\n");
11498:
11499: fflush(ficlog);
11500:
11501: } /* End of mle != -3 */
1.218 brouard 11502:
1.186 brouard 11503: /* Main data
11504: */
1.290 brouard 11505: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11506: /* num=lvector(1,n); */
11507: /* moisnais=vector(1,n); */
11508: /* annais=vector(1,n); */
11509: /* moisdc=vector(1,n); */
11510: /* andc=vector(1,n); */
11511: /* weight=vector(1,n); */
11512: /* agedc=vector(1,n); */
11513: /* cod=ivector(1,n); */
11514: /* for(i=1;i<=n;i++){ */
11515: num=lvector(firstobs,lastobs);
11516: moisnais=vector(firstobs,lastobs);
11517: annais=vector(firstobs,lastobs);
11518: moisdc=vector(firstobs,lastobs);
11519: andc=vector(firstobs,lastobs);
11520: weight=vector(firstobs,lastobs);
11521: agedc=vector(firstobs,lastobs);
11522: cod=ivector(firstobs,lastobs);
11523: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11524: num[i]=0;
11525: moisnais[i]=0;
11526: annais[i]=0;
11527: moisdc[i]=0;
11528: andc[i]=0;
11529: agedc[i]=0;
11530: cod[i]=0;
11531: weight[i]=1.0; /* Equal weights, 1 by default */
11532: }
1.290 brouard 11533: mint=matrix(1,maxwav,firstobs,lastobs);
11534: anint=matrix(1,maxwav,firstobs,lastobs);
11535: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11536: tab=ivector(1,NCOVMAX);
1.144 brouard 11537: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11538: 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 11539:
1.136 brouard 11540: /* Reads data from file datafile */
11541: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11542: goto end;
11543:
11544: /* Calculation of the number of parameters from char model */
1.234 brouard 11545: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11546: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11547: k=3 V4 Tvar[k=3]= 4 (from V4)
11548: k=2 V1 Tvar[k=2]= 1 (from V1)
11549: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11550: */
11551:
11552: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11553: TvarsDind=ivector(1,NCOVMAX); /* */
11554: TvarsD=ivector(1,NCOVMAX); /* */
11555: TvarsQind=ivector(1,NCOVMAX); /* */
11556: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11557: TvarF=ivector(1,NCOVMAX); /* */
11558: TvarFind=ivector(1,NCOVMAX); /* */
11559: TvarV=ivector(1,NCOVMAX); /* */
11560: TvarVind=ivector(1,NCOVMAX); /* */
11561: TvarA=ivector(1,NCOVMAX); /* */
11562: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11563: TvarFD=ivector(1,NCOVMAX); /* */
11564: TvarFDind=ivector(1,NCOVMAX); /* */
11565: TvarFQ=ivector(1,NCOVMAX); /* */
11566: TvarFQind=ivector(1,NCOVMAX); /* */
11567: TvarVD=ivector(1,NCOVMAX); /* */
11568: TvarVDind=ivector(1,NCOVMAX); /* */
11569: TvarVQ=ivector(1,NCOVMAX); /* */
11570: TvarVQind=ivector(1,NCOVMAX); /* */
11571:
1.230 brouard 11572: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11573: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11574: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11575: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11576: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11577: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11578: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11579: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11580: */
11581: /* For model-covariate k tells which data-covariate to use but
11582: because this model-covariate is a construction we invent a new column
11583: ncovcol + k1
11584: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11585: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11586: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11587: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11588: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11589: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11590: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11591: */
1.145 brouard 11592: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11593: 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 11594: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11595: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11596: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11597: 4 covariates (3 plus signs)
11598: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11599: */
1.230 brouard 11600: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11601: * individual dummy, fixed or varying:
11602: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11603: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11604: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11605: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11606: * Tmodelind[1]@9={9,0,3,2,}*/
11607: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11608: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11609: * individual quantitative, fixed or varying:
11610: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11611: * 3, 1, 0, 0, 0, 0, 0, 0},
11612: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11613: /* Main decodemodel */
11614:
1.187 brouard 11615:
1.223 brouard 11616: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11617: goto end;
11618:
1.137 brouard 11619: if((double)(lastobs-imx)/(double)imx > 1.10){
11620: nbwarn++;
11621: 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);
11622: 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);
11623: }
1.136 brouard 11624: /* if(mle==1){*/
1.137 brouard 11625: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11626: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11627: }
11628:
11629: /*-calculation of age at interview from date of interview and age at death -*/
11630: agev=matrix(1,maxwav,1,imx);
11631:
11632: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11633: goto end;
11634:
1.126 brouard 11635:
1.136 brouard 11636: agegomp=(int)agemin;
1.290 brouard 11637: free_vector(moisnais,firstobs,lastobs);
11638: free_vector(annais,firstobs,lastobs);
1.126 brouard 11639: /* free_matrix(mint,1,maxwav,1,n);
11640: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11641: /* free_vector(moisdc,1,n); */
11642: /* free_vector(andc,1,n); */
1.145 brouard 11643: /* */
11644:
1.126 brouard 11645: wav=ivector(1,imx);
1.214 brouard 11646: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11647: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11648: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11649: 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.*/
11650: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11651: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11652:
11653: /* Concatenates waves */
1.214 brouard 11654: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11655: Death is a valid wave (if date is known).
11656: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11657: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11658: and mw[mi+1][i]. dh depends on stepm.
11659: */
11660:
1.126 brouard 11661: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11662: /* Concatenates waves */
1.145 brouard 11663:
1.290 brouard 11664: free_vector(moisdc,firstobs,lastobs);
11665: free_vector(andc,firstobs,lastobs);
1.215 brouard 11666:
1.126 brouard 11667: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11668: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11669: ncodemax[1]=1;
1.145 brouard 11670: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11671: cptcoveff=0;
1.220 brouard 11672: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11673: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11674: }
11675:
11676: ncovcombmax=pow(2,cptcoveff);
11677: invalidvarcomb=ivector(1, ncovcombmax);
11678: for(i=1;i<ncovcombmax;i++)
11679: invalidvarcomb[i]=0;
11680:
1.211 brouard 11681: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11682: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11683: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11684:
1.200 brouard 11685: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11686: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11687: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11688: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11689: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11690: * (currently 0 or 1) in the data.
11691: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11692: * corresponding modality (h,j).
11693: */
11694:
1.145 brouard 11695: h=0;
11696: /*if (cptcovn > 0) */
1.126 brouard 11697: m=pow(2,cptcoveff);
11698:
1.144 brouard 11699: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11700: * For k=4 covariates, h goes from 1 to m=2**k
11701: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11702: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11703: * h\k 1 2 3 4
1.143 brouard 11704: *______________________________
11705: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11706: * 2 2 1 1 1
11707: * 3 i=2 1 2 1 1
11708: * 4 2 2 1 1
11709: * 5 i=3 1 i=2 1 2 1
11710: * 6 2 1 2 1
11711: * 7 i=4 1 2 2 1
11712: * 8 2 2 2 1
1.197 brouard 11713: * 9 i=5 1 i=3 1 i=2 1 2
11714: * 10 2 1 1 2
11715: * 11 i=6 1 2 1 2
11716: * 12 2 2 1 2
11717: * 13 i=7 1 i=4 1 2 2
11718: * 14 2 1 2 2
11719: * 15 i=8 1 2 2 2
11720: * 16 2 2 2 2
1.143 brouard 11721: */
1.212 brouard 11722: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11723: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11724: * and the value of each covariate?
11725: * V1=1, V2=1, V3=2, V4=1 ?
11726: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11727: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11728: * In order to get the real value in the data, we use nbcode
11729: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11730: * We are keeping this crazy system in order to be able (in the future?)
11731: * to have more than 2 values (0 or 1) for a covariate.
11732: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11733: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11734: * bbbbbbbb
11735: * 76543210
11736: * h-1 00000101 (6-1=5)
1.219 brouard 11737: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11738: * &
11739: * 1 00000001 (1)
1.219 brouard 11740: * 00000000 = 1 & ((h-1) >> (k-1))
11741: * +1= 00000001 =1
1.211 brouard 11742: *
11743: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11744: * h' 1101 =2^3+2^2+0x2^1+2^0
11745: * >>k' 11
11746: * & 00000001
11747: * = 00000001
11748: * +1 = 00000010=2 = codtabm(14,3)
11749: * Reverse h=6 and m=16?
11750: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11751: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11752: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11753: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11754: * V3=decodtabm(14,3,2**4)=2
11755: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11756: *(h-1) >> (j-1) 0011 =13 >> 2
11757: * &1 000000001
11758: * = 000000001
11759: * +1= 000000010 =2
11760: * 2211
11761: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11762: * V3=2
1.220 brouard 11763: * codtabm and decodtabm are identical
1.211 brouard 11764: */
11765:
1.145 brouard 11766:
11767: free_ivector(Ndum,-1,NCOVMAX);
11768:
11769:
1.126 brouard 11770:
1.186 brouard 11771: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11772: strcpy(optionfilegnuplot,optionfilefiname);
11773: if(mle==-3)
1.201 brouard 11774: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11775: strcat(optionfilegnuplot,".gp");
11776:
11777: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11778: printf("Problem with file %s",optionfilegnuplot);
11779: }
11780: else{
1.204 brouard 11781: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11782: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11783: //fprintf(ficgp,"set missing 'NaNq'\n");
11784: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11785: }
11786: /* fclose(ficgp);*/
1.186 brouard 11787:
11788:
11789: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11790:
11791: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11792: if(mle==-3)
1.201 brouard 11793: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11794: strcat(optionfilehtm,".htm");
11795: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11796: printf("Problem with %s \n",optionfilehtm);
11797: exit(0);
1.126 brouard 11798: }
11799:
11800: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11801: strcat(optionfilehtmcov,"-cov.htm");
11802: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11803: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11804: }
11805: else{
11806: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11807: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11808: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11809: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11810: }
11811:
1.213 brouard 11812: 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 11813: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11814: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11815: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11816: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11817: \n\
11818: <hr size=\"2\" color=\"#EC5E5E\">\
11819: <ul><li><h4>Parameter files</h4>\n\
11820: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11821: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11822: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11823: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11824: - Date and time at start: %s</ul>\n",\
11825: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11826: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11827: fileres,fileres,\
11828: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11829: fflush(fichtm);
11830:
11831: strcpy(pathr,path);
11832: strcat(pathr,optionfilefiname);
1.184 brouard 11833: #ifdef WIN32
11834: _chdir(optionfilefiname); /* Move to directory named optionfile */
11835: #else
1.126 brouard 11836: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11837: #endif
11838:
1.126 brouard 11839:
1.220 brouard 11840: /* Calculates basic frequencies. Computes observed prevalence at single age
11841: and for any valid combination of covariates
1.126 brouard 11842: and prints on file fileres'p'. */
1.251 brouard 11843: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11844: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11845:
11846: fprintf(fichtm,"\n");
1.286 brouard 11847: fprintf(fichtm,"<h4>Parameter line 2</h4><ul><li>Tolerance for the convergence of the likelihood: ftol=%g \n<li>Interval for the elementary matrix (in month): stepm=%d",\
1.274 brouard 11848: ftol, stepm);
11849: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11850: ncurrv=1;
11851: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11852: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11853: ncurrv=i;
11854: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11855: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11856: ncurrv=i;
11857: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11858: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11859: ncurrv=i;
11860: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11861: fprintf(fichtm,"\n<li>Weights column \n<br>Number of alive states: nlstate=%d <br>Number of death states (not really implemented): ndeath=%d \n<li>Number of waves: maxwav=%d \n<li>Parameter for maximization (1), using parameter values (0), for design of parameters and variance-covariance matrix: mle=%d \n<li>Does the weight column be taken into account (1), or not (0): weight=%d</ul>\n", \
11862: nlstate, ndeath, maxwav, mle, weightopt);
11863:
11864: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11865: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11866:
11867:
11868: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11869: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11870: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11871: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11872: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11873: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11874: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11875: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11876: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11877:
1.126 brouard 11878: /* For Powell, parameters are in a vector p[] starting at p[1]
11879: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11880: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11881:
11882: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11883: /* For mortality only */
1.126 brouard 11884: if (mle==-3){
1.136 brouard 11885: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11886: for(i=1;i<=NDIM;i++)
11887: for(j=1;j<=NDIM;j++)
11888: ximort[i][j]=0.;
1.186 brouard 11889: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11890: cens=ivector(firstobs,lastobs);
11891: ageexmed=vector(firstobs,lastobs);
11892: agecens=vector(firstobs,lastobs);
11893: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11894:
1.126 brouard 11895: for (i=1; i<=imx; i++){
11896: dcwave[i]=-1;
11897: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11898: if (s[m][i]>nlstate) {
11899: dcwave[i]=m;
11900: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11901: break;
11902: }
1.126 brouard 11903: }
1.226 brouard 11904:
1.126 brouard 11905: for (i=1; i<=imx; i++) {
11906: if (wav[i]>0){
1.226 brouard 11907: ageexmed[i]=agev[mw[1][i]][i];
11908: j=wav[i];
11909: agecens[i]=1.;
11910:
11911: if (ageexmed[i]> 1 && wav[i] > 0){
11912: agecens[i]=agev[mw[j][i]][i];
11913: cens[i]= 1;
11914: }else if (ageexmed[i]< 1)
11915: cens[i]= -1;
11916: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11917: cens[i]=0 ;
1.126 brouard 11918: }
11919: else cens[i]=-1;
11920: }
11921:
11922: for (i=1;i<=NDIM;i++) {
11923: for (j=1;j<=NDIM;j++)
1.226 brouard 11924: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11925: }
11926:
1.302 ! brouard 11927: p[1]=0.0268; p[NDIM]=0.083;
! 11928: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 11929:
11930:
1.136 brouard 11931: #ifdef GSL
11932: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11933: #else
1.126 brouard 11934: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11935: #endif
1.201 brouard 11936: strcpy(filerespow,"POW-MORT_");
11937: strcat(filerespow,fileresu);
1.126 brouard 11938: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11939: printf("Problem with resultfile: %s\n", filerespow);
11940: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11941: }
1.136 brouard 11942: #ifdef GSL
11943: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11944: #else
1.126 brouard 11945: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11946: #endif
1.126 brouard 11947: /* for (i=1;i<=nlstate;i++)
11948: for(j=1;j<=nlstate+ndeath;j++)
11949: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11950: */
11951: fprintf(ficrespow,"\n");
1.136 brouard 11952: #ifdef GSL
11953: /* gsl starts here */
11954: T = gsl_multimin_fminimizer_nmsimplex;
11955: gsl_multimin_fminimizer *sfm = NULL;
11956: gsl_vector *ss, *x;
11957: gsl_multimin_function minex_func;
11958:
11959: /* Initial vertex size vector */
11960: ss = gsl_vector_alloc (NDIM);
11961:
11962: if (ss == NULL){
11963: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11964: }
11965: /* Set all step sizes to 1 */
11966: gsl_vector_set_all (ss, 0.001);
11967:
11968: /* Starting point */
1.126 brouard 11969:
1.136 brouard 11970: x = gsl_vector_alloc (NDIM);
11971:
11972: if (x == NULL){
11973: gsl_vector_free(ss);
11974: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11975: }
11976:
11977: /* Initialize method and iterate */
11978: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11979: /* gsl_vector_set(x, 0, 0.0268); */
11980: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11981: gsl_vector_set(x, 0, p[1]);
11982: gsl_vector_set(x, 1, p[2]);
11983:
11984: minex_func.f = &gompertz_f;
11985: minex_func.n = NDIM;
11986: minex_func.params = (void *)&p; /* ??? */
11987:
11988: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11989: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11990:
11991: printf("Iterations beginning .....\n\n");
11992: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11993:
11994: iteri=0;
11995: while (rval == GSL_CONTINUE){
11996: iteri++;
11997: status = gsl_multimin_fminimizer_iterate(sfm);
11998:
11999: if (status) printf("error: %s\n", gsl_strerror (status));
12000: fflush(0);
12001:
12002: if (status)
12003: break;
12004:
12005: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12006: ssval = gsl_multimin_fminimizer_size (sfm);
12007:
12008: if (rval == GSL_SUCCESS)
12009: printf ("converged to a local maximum at\n");
12010:
12011: printf("%5d ", iteri);
12012: for (it = 0; it < NDIM; it++){
12013: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12014: }
12015: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12016: }
12017:
12018: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12019:
12020: gsl_vector_free(x); /* initial values */
12021: gsl_vector_free(ss); /* inital step size */
12022: for (it=0; it<NDIM; it++){
12023: p[it+1]=gsl_vector_get(sfm->x,it);
12024: fprintf(ficrespow," %.12lf", p[it]);
12025: }
12026: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12027: #endif
12028: #ifdef POWELL
12029: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12030: #endif
1.126 brouard 12031: fclose(ficrespow);
12032:
1.203 brouard 12033: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12034:
12035: for(i=1; i <=NDIM; i++)
12036: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12037: matcov[i][j]=matcov[j][i];
1.126 brouard 12038:
12039: printf("\nCovariance matrix\n ");
1.203 brouard 12040: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12041: for(i=1; i <=NDIM; i++) {
12042: for(j=1;j<=NDIM;j++){
1.220 brouard 12043: printf("%f ",matcov[i][j]);
12044: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12045: }
1.203 brouard 12046: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12047: }
12048:
12049: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12050: for (i=1;i<=NDIM;i++) {
1.126 brouard 12051: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12052: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12053: }
1.302 ! brouard 12054: lsurv=vector(agegomp,AGESUP);
! 12055: lpop=vector(agegomp,AGESUP);
! 12056: tpop=vector(agegomp,AGESUP);
1.126 brouard 12057: lsurv[agegomp]=100000;
12058:
12059: for (k=agegomp;k<=AGESUP;k++) {
12060: agemortsup=k;
12061: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12062: }
12063:
12064: for (k=agegomp;k<agemortsup;k++)
12065: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12066:
12067: for (k=agegomp;k<agemortsup;k++){
12068: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12069: sumlpop=sumlpop+lpop[k];
12070: }
12071:
12072: tpop[agegomp]=sumlpop;
12073: for (k=agegomp;k<(agemortsup-3);k++){
12074: /* tpop[k+1]=2;*/
12075: tpop[k+1]=tpop[k]-lpop[k];
12076: }
12077:
12078:
12079: printf("\nAge lx qx dx Lx Tx e(x)\n");
12080: for (k=agegomp;k<(agemortsup-2);k++)
12081: 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]);
12082:
12083:
12084: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12085: ageminpar=50;
12086: agemaxpar=100;
1.194 brouard 12087: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12088: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12089: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12090: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12091: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12092: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12093: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12094: }else{
12095: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12096: 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 12097: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12098: }
1.201 brouard 12099: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12100: stepm, weightopt,\
12101: model,imx,p,matcov,agemortsup);
12102:
1.302 ! brouard 12103: free_vector(lsurv,agegomp,AGESUP);
! 12104: free_vector(lpop,agegomp,AGESUP);
! 12105: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12106: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12107: free_ivector(dcwave,firstobs,lastobs);
12108: free_vector(agecens,firstobs,lastobs);
12109: free_vector(ageexmed,firstobs,lastobs);
12110: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12111: #ifdef GSL
1.136 brouard 12112: #endif
1.186 brouard 12113: } /* Endof if mle==-3 mortality only */
1.205 brouard 12114: /* Standard */
12115: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12116: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12117: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12118: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12119: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12120: for (k=1; k<=npar;k++)
12121: printf(" %d %8.5f",k,p[k]);
12122: printf("\n");
1.205 brouard 12123: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12124: /* mlikeli uses func not funcone */
1.247 brouard 12125: /* for(i=1;i<nlstate;i++){ */
12126: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12127: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12128: /* } */
1.205 brouard 12129: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12130: }
12131: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12132: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12133: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12134: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12135: }
12136: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12137: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12138: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12139: for (k=1; k<=npar;k++)
12140: printf(" %d %8.5f",k,p[k]);
12141: printf("\n");
12142:
12143: /*--------- results files --------------*/
1.283 brouard 12144: /* 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 12145:
12146:
12147: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12148: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12149: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12150: for(i=1,jk=1; i <=nlstate; i++){
12151: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12152: if (k != i) {
12153: printf("%d%d ",i,k);
12154: fprintf(ficlog,"%d%d ",i,k);
12155: fprintf(ficres,"%1d%1d ",i,k);
12156: for(j=1; j <=ncovmodel; j++){
12157: printf("%12.7f ",p[jk]);
12158: fprintf(ficlog,"%12.7f ",p[jk]);
12159: fprintf(ficres,"%12.7f ",p[jk]);
12160: jk++;
12161: }
12162: printf("\n");
12163: fprintf(ficlog,"\n");
12164: fprintf(ficres,"\n");
12165: }
1.126 brouard 12166: }
12167: }
1.203 brouard 12168: if(mle != 0){
12169: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12170: ftolhess=ftol; /* Usually correct */
1.203 brouard 12171: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12172: 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");
12173: 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");
12174: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12175: for(k=1; k <=(nlstate+ndeath); k++){
12176: if (k != i) {
12177: printf("%d%d ",i,k);
12178: fprintf(ficlog,"%d%d ",i,k);
12179: for(j=1; j <=ncovmodel; j++){
12180: 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]));
12181: 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]));
12182: jk++;
12183: }
12184: printf("\n");
12185: fprintf(ficlog,"\n");
12186: }
12187: }
1.193 brouard 12188: }
1.203 brouard 12189: } /* end of hesscov and Wald tests */
1.225 brouard 12190:
1.203 brouard 12191: /* */
1.126 brouard 12192: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12193: printf("# Scales (for hessian or gradient estimation)\n");
12194: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12195: for(i=1,jk=1; i <=nlstate; i++){
12196: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12197: if (j!=i) {
12198: fprintf(ficres,"%1d%1d",i,j);
12199: printf("%1d%1d",i,j);
12200: fprintf(ficlog,"%1d%1d",i,j);
12201: for(k=1; k<=ncovmodel;k++){
12202: printf(" %.5e",delti[jk]);
12203: fprintf(ficlog," %.5e",delti[jk]);
12204: fprintf(ficres," %.5e",delti[jk]);
12205: jk++;
12206: }
12207: printf("\n");
12208: fprintf(ficlog,"\n");
12209: fprintf(ficres,"\n");
12210: }
1.126 brouard 12211: }
12212: }
12213:
12214: 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 12215: if(mle >= 1) /* To big for the screen */
1.126 brouard 12216: 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");
12217: 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");
12218: /* # 121 Var(a12)\n\ */
12219: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12220: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12221: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12222: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12223: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12224: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12225: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12226:
12227:
12228: /* Just to have a covariance matrix which will be more understandable
12229: even is we still don't want to manage dictionary of variables
12230: */
12231: for(itimes=1;itimes<=2;itimes++){
12232: jj=0;
12233: for(i=1; i <=nlstate; i++){
1.225 brouard 12234: for(j=1; j <=nlstate+ndeath; j++){
12235: if(j==i) continue;
12236: for(k=1; k<=ncovmodel;k++){
12237: jj++;
12238: ca[0]= k+'a'-1;ca[1]='\0';
12239: if(itimes==1){
12240: if(mle>=1)
12241: printf("#%1d%1d%d",i,j,k);
12242: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12243: fprintf(ficres,"#%1d%1d%d",i,j,k);
12244: }else{
12245: if(mle>=1)
12246: printf("%1d%1d%d",i,j,k);
12247: fprintf(ficlog,"%1d%1d%d",i,j,k);
12248: fprintf(ficres,"%1d%1d%d",i,j,k);
12249: }
12250: ll=0;
12251: for(li=1;li <=nlstate; li++){
12252: for(lj=1;lj <=nlstate+ndeath; lj++){
12253: if(lj==li) continue;
12254: for(lk=1;lk<=ncovmodel;lk++){
12255: ll++;
12256: if(ll<=jj){
12257: cb[0]= lk +'a'-1;cb[1]='\0';
12258: if(ll<jj){
12259: if(itimes==1){
12260: if(mle>=1)
12261: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12262: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12263: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12264: }else{
12265: if(mle>=1)
12266: printf(" %.5e",matcov[jj][ll]);
12267: fprintf(ficlog," %.5e",matcov[jj][ll]);
12268: fprintf(ficres," %.5e",matcov[jj][ll]);
12269: }
12270: }else{
12271: if(itimes==1){
12272: if(mle>=1)
12273: printf(" Var(%s%1d%1d)",ca,i,j);
12274: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12275: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12276: }else{
12277: if(mle>=1)
12278: printf(" %.7e",matcov[jj][ll]);
12279: fprintf(ficlog," %.7e",matcov[jj][ll]);
12280: fprintf(ficres," %.7e",matcov[jj][ll]);
12281: }
12282: }
12283: }
12284: } /* end lk */
12285: } /* end lj */
12286: } /* end li */
12287: if(mle>=1)
12288: printf("\n");
12289: fprintf(ficlog,"\n");
12290: fprintf(ficres,"\n");
12291: numlinepar++;
12292: } /* end k*/
12293: } /*end j */
1.126 brouard 12294: } /* end i */
12295: } /* end itimes */
12296:
12297: fflush(ficlog);
12298: fflush(ficres);
1.225 brouard 12299: while(fgets(line, MAXLINE, ficpar)) {
12300: /* If line starts with a # it is a comment */
12301: if (line[0] == '#') {
12302: numlinepar++;
12303: fputs(line,stdout);
12304: fputs(line,ficparo);
12305: fputs(line,ficlog);
1.299 brouard 12306: fputs(line,ficres);
1.225 brouard 12307: continue;
12308: }else
12309: break;
12310: }
12311:
1.209 brouard 12312: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12313: /* ungetc(c,ficpar); */
12314: /* fgets(line, MAXLINE, ficpar); */
12315: /* fputs(line,stdout); */
12316: /* fputs(line,ficparo); */
12317: /* } */
12318: /* ungetc(c,ficpar); */
1.126 brouard 12319:
12320: estepm=0;
1.209 brouard 12321: 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 12322:
12323: if (num_filled != 6) {
12324: 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);
12325: 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);
12326: goto end;
12327: }
12328: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12329: }
12330: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12331: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12332:
1.209 brouard 12333: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12334: if (estepm==0 || estepm < stepm) estepm=stepm;
12335: if (fage <= 2) {
12336: bage = ageminpar;
12337: fage = agemaxpar;
12338: }
12339:
12340: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12341: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12342: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12343:
1.186 brouard 12344: /* Other stuffs, more or less useful */
1.254 brouard 12345: while(fgets(line, MAXLINE, ficpar)) {
12346: /* If line starts with a # it is a comment */
12347: if (line[0] == '#') {
12348: numlinepar++;
12349: fputs(line,stdout);
12350: fputs(line,ficparo);
12351: fputs(line,ficlog);
1.299 brouard 12352: fputs(line,ficres);
1.254 brouard 12353: continue;
12354: }else
12355: break;
12356: }
12357:
12358: 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){
12359:
12360: if (num_filled != 7) {
12361: 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);
12362: 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);
12363: goto end;
12364: }
12365: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12366: 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);
12367: 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);
12368: 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 12369: }
1.254 brouard 12370:
12371: while(fgets(line, MAXLINE, ficpar)) {
12372: /* If line starts with a # it is a comment */
12373: if (line[0] == '#') {
12374: numlinepar++;
12375: fputs(line,stdout);
12376: fputs(line,ficparo);
12377: fputs(line,ficlog);
1.299 brouard 12378: fputs(line,ficres);
1.254 brouard 12379: continue;
12380: }else
12381: break;
1.126 brouard 12382: }
12383:
12384:
12385: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12386: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12387:
1.254 brouard 12388: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12389: if (num_filled != 1) {
12390: 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);
12391: 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);
12392: goto end;
12393: }
12394: printf("pop_based=%d\n",popbased);
12395: fprintf(ficlog,"pop_based=%d\n",popbased);
12396: fprintf(ficparo,"pop_based=%d\n",popbased);
12397: fprintf(ficres,"pop_based=%d\n",popbased);
12398: }
12399:
1.258 brouard 12400: /* Results */
12401: nresult=0;
12402: do{
12403: if(!fgets(line, MAXLINE, ficpar)){
12404: endishere=1;
12405: parameterline=14;
12406: }else if (line[0] == '#') {
12407: /* If line starts with a # it is a comment */
1.254 brouard 12408: numlinepar++;
12409: fputs(line,stdout);
12410: fputs(line,ficparo);
12411: fputs(line,ficlog);
1.299 brouard 12412: fputs(line,ficres);
1.254 brouard 12413: continue;
1.258 brouard 12414: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12415: parameterline=11;
1.296 brouard 12416: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12417: parameterline=12;
12418: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12419: parameterline=13;
12420: else{
12421: parameterline=14;
1.254 brouard 12422: }
1.258 brouard 12423: switch (parameterline){
12424: case 11:
1.296 brouard 12425: 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 && (num_filled == 8)){
12426: 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);
1.258 brouard 12427: 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);
12428: 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);
12429: 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);
12430: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12431: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12432: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12433: prvforecast = 1;
12434: }
12435: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.302 ! brouard 12436: printf("prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
! 12437: fprintf(ficlog,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
! 12438: fprintf(ficres,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12439: prvforecast = 2;
12440: }
12441: else {
12442: 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\nnor 3 (data)parameters, for example:prevforecast=1 yearsfproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12443: 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 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12444: goto end;
1.258 brouard 12445: }
1.254 brouard 12446: break;
1.258 brouard 12447: case 12:
1.296 brouard 12448: if((num_filled=sscanf(line,"prevbackcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&prevbcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF && (num_filled == 8)){
12449: fprintf(ficparo,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12450: printf("prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12451: fprintf(ficlog,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12452: fprintf(ficres,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12453: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12454: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12455: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12456: prvbackcast = 1;
12457: }
12458: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.302 ! brouard 12459: printf("prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
! 12460: fprintf(ficlog,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
! 12461: fprintf(ficres,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12462: prvbackcast = 2;
12463: }
12464: else {
12465: printf("Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearsbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12466: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12467: goto end;
1.258 brouard 12468: }
1.230 brouard 12469: break;
1.258 brouard 12470: case 13:
12471: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12472: if (num_filled == 0){
12473: resultline[0]='\0';
12474: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12475: 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);
12476: break;
12477: } else if (num_filled != 1){
12478: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12479: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12480: }
12481: nresult++; /* Sum of resultlines */
12482: printf("Result %d: result=%s\n",nresult, resultline);
12483: if(nresult > MAXRESULTLINES){
12484: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12485: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12486: goto end;
12487: }
12488: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12489: fprintf(ficparo,"result: %s\n",resultline);
12490: fprintf(ficres,"result: %s\n",resultline);
12491: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12492: break;
1.258 brouard 12493: case 14:
1.259 brouard 12494: if(ncovmodel >2 && nresult==0 ){
12495: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12496: goto end;
12497: }
1.259 brouard 12498: break;
1.258 brouard 12499: default:
12500: nresult=1;
12501: decoderesult(".",nresult ); /* No covariate */
12502: }
12503: } /* End switch parameterline */
12504: }while(endishere==0); /* End do */
1.126 brouard 12505:
1.230 brouard 12506: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12507: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12508:
12509: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12510: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12511: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12512: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12513: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12514: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12515: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12516: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12517: }else{
1.270 brouard 12518: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12519: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12520: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12521: if(prvforecast==1){
12522: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12523: jprojd=jproj1;
12524: mprojd=mproj1;
12525: anprojd=anproj1;
12526: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12527: jprojf=jproj2;
12528: mprojf=mproj2;
12529: anprojf=anproj2;
12530: } else if(prvforecast == 2){
12531: dateprojd=dateintmean;
12532: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12533: dateprojf=dateintmean+yrfproj;
12534: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12535: }
12536: if(prvbackcast==1){
12537: datebackd=(jback1+12*mback1+365*anback1)/365;
12538: jbackd=jback1;
12539: mbackd=mback1;
12540: anbackd=anback1;
12541: datebackf=(jback2+12*mback2+365*anback2)/365;
12542: jbackf=jback2;
12543: mbackf=mback2;
12544: anbackf=anback2;
12545: } else if(prvbackcast == 2){
12546: datebackd=dateintmean;
12547: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12548: datebackf=dateintmean-yrbproj;
12549: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12550: }
12551:
12552: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12553: }
12554: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12555: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12556: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12557:
1.225 brouard 12558: /*------------ free_vector -------------*/
12559: /* chdir(path); */
1.220 brouard 12560:
1.215 brouard 12561: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12562: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12563: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12564: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12565: free_lvector(num,firstobs,lastobs);
12566: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12567: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12568: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12569: fclose(ficparo);
12570: fclose(ficres);
1.220 brouard 12571:
12572:
1.186 brouard 12573: /* Other results (useful)*/
1.220 brouard 12574:
12575:
1.126 brouard 12576: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12577: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12578: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12579: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12580: fclose(ficrespl);
12581:
12582: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12583: /*#include "hpijx.h"*/
12584: hPijx(p, bage, fage);
1.145 brouard 12585: fclose(ficrespij);
1.227 brouard 12586:
1.220 brouard 12587: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12588: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12589: k=1;
1.126 brouard 12590: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12591:
1.269 brouard 12592: /* Prevalence for each covariate combination in probs[age][status][cov] */
12593: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12594: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12595: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12596: for(k=1;k<=ncovcombmax;k++)
12597: probs[i][j][k]=0.;
1.269 brouard 12598: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12599: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12600: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12601: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12602: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12603: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12604: for(k=1;k<=ncovcombmax;k++)
12605: mobaverages[i][j][k]=0.;
1.219 brouard 12606: mobaverage=mobaverages;
12607: if (mobilav!=0) {
1.235 brouard 12608: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12609: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12610: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12611: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12612: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12613: }
1.269 brouard 12614: } else if (mobilavproj !=0) {
1.235 brouard 12615: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12616: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12617: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12618: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12619: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12620: }
1.269 brouard 12621: }else{
12622: printf("Internal error moving average\n");
12623: fflush(stdout);
12624: exit(1);
1.219 brouard 12625: }
12626: }/* end if moving average */
1.227 brouard 12627:
1.126 brouard 12628: /*---------- Forecasting ------------------*/
1.296 brouard 12629: if(prevfcast==1){
12630: /* /\* if(stepm ==1){*\/ */
12631: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12632: /*This done previously after freqsummary.*/
12633: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12634: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12635:
12636: /* } else if (prvforecast==2){ */
12637: /* /\* if(stepm ==1){*\/ */
12638: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12639: /* } */
12640: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12641: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12642: }
1.269 brouard 12643:
1.296 brouard 12644: /* Prevbcasting */
12645: if(prevbcast==1){
1.219 brouard 12646: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12647: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12648: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12649:
12650: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12651:
12652: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12653:
1.219 brouard 12654: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12655: fclose(ficresplb);
12656:
1.222 brouard 12657: hBijx(p, bage, fage, mobaverage);
12658: fclose(ficrespijb);
1.219 brouard 12659:
1.296 brouard 12660: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12661: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12662: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12663: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12664: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12665: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12666:
12667:
1.269 brouard 12668: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12669:
12670:
1.269 brouard 12671: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12672: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12673: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12674: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12675: } /* end Prevbcasting */
1.268 brouard 12676:
1.186 brouard 12677:
12678: /* ------ Other prevalence ratios------------ */
1.126 brouard 12679:
1.215 brouard 12680: free_ivector(wav,1,imx);
12681: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12682: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12683: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12684:
12685:
1.127 brouard 12686: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12687:
1.201 brouard 12688: strcpy(filerese,"E_");
12689: strcat(filerese,fileresu);
1.126 brouard 12690: if((ficreseij=fopen(filerese,"w"))==NULL) {
12691: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12692: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12693: }
1.208 brouard 12694: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12695: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12696:
12697: pstamp(ficreseij);
1.219 brouard 12698:
1.235 brouard 12699: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12700: if (cptcovn < 1){i1=1;}
12701:
12702: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12703: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12704: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12705: continue;
1.219 brouard 12706: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12707: printf("\n#****** ");
1.225 brouard 12708: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12709: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12710: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12711: }
12712: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12713: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12714: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12715: }
12716: fprintf(ficreseij,"******\n");
1.235 brouard 12717: printf("******\n");
1.219 brouard 12718:
12719: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12720: oldm=oldms;savm=savms;
1.235 brouard 12721: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12722:
1.219 brouard 12723: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12724: }
12725: fclose(ficreseij);
1.208 brouard 12726: printf("done evsij\n");fflush(stdout);
12727: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12728:
1.218 brouard 12729:
1.227 brouard 12730: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12731:
1.201 brouard 12732: strcpy(filerest,"T_");
12733: strcat(filerest,fileresu);
1.127 brouard 12734: if((ficrest=fopen(filerest,"w"))==NULL) {
12735: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12736: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12737: }
1.208 brouard 12738: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12739: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12740: strcpy(fileresstde,"STDE_");
12741: strcat(fileresstde,fileresu);
1.126 brouard 12742: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12743: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12744: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12745: }
1.227 brouard 12746: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12747: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12748:
1.201 brouard 12749: strcpy(filerescve,"CVE_");
12750: strcat(filerescve,fileresu);
1.126 brouard 12751: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12752: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12753: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12754: }
1.227 brouard 12755: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12756: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12757:
1.201 brouard 12758: strcpy(fileresv,"V_");
12759: strcat(fileresv,fileresu);
1.126 brouard 12760: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12761: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12762: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12763: }
1.227 brouard 12764: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12765: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12766:
1.235 brouard 12767: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12768: if (cptcovn < 1){i1=1;}
12769:
12770: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12771: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12772: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12773: continue;
1.242 brouard 12774: printf("\n#****** Result for:");
12775: fprintf(ficrest,"\n#****** Result for:");
12776: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12777: for(j=1;j<=cptcoveff;j++){
12778: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12779: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12780: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12781: }
1.235 brouard 12782: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12783: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12784: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12785: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12786: }
1.208 brouard 12787: fprintf(ficrest,"******\n");
1.227 brouard 12788: fprintf(ficlog,"******\n");
12789: printf("******\n");
1.208 brouard 12790:
12791: fprintf(ficresstdeij,"\n#****** ");
12792: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12793: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12794: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12795: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12796: }
1.235 brouard 12797: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12798: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12799: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12800: }
1.208 brouard 12801: fprintf(ficresstdeij,"******\n");
12802: fprintf(ficrescveij,"******\n");
12803:
12804: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12805: /* pstamp(ficresvij); */
1.225 brouard 12806: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12807: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12808: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12809: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12810: }
1.208 brouard 12811: fprintf(ficresvij,"******\n");
12812:
12813: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12814: oldm=oldms;savm=savms;
1.235 brouard 12815: printf(" cvevsij ");
12816: fprintf(ficlog, " cvevsij ");
12817: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12818: printf(" end cvevsij \n ");
12819: fprintf(ficlog, " end cvevsij \n ");
12820:
12821: /*
12822: */
12823: /* goto endfree; */
12824:
12825: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12826: pstamp(ficrest);
12827:
1.269 brouard 12828: epj=vector(1,nlstate+1);
1.208 brouard 12829: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12830: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12831: cptcod= 0; /* To be deleted */
12832: printf("varevsij vpopbased=%d \n",vpopbased);
12833: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12834: 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 12835: 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 ");
12836: if(vpopbased==1)
12837: 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);
12838: else
1.288 brouard 12839: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12840: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12841: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12842: fprintf(ficrest,"\n");
12843: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12844: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12845: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12846: for(age=bage; age <=fage ;age++){
1.235 brouard 12847: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12848: if (vpopbased==1) {
12849: if(mobilav ==0){
12850: for(i=1; i<=nlstate;i++)
12851: prlim[i][i]=probs[(int)age][i][k];
12852: }else{ /* mobilav */
12853: for(i=1; i<=nlstate;i++)
12854: prlim[i][i]=mobaverage[(int)age][i][k];
12855: }
12856: }
1.219 brouard 12857:
1.227 brouard 12858: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12859: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12860: /* printf(" age %4.0f ",age); */
12861: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12862: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12863: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12864: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12865: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12866: }
12867: epj[nlstate+1] +=epj[j];
12868: }
12869: /* printf(" age %4.0f \n",age); */
1.219 brouard 12870:
1.227 brouard 12871: for(i=1, vepp=0.;i <=nlstate;i++)
12872: for(j=1;j <=nlstate;j++)
12873: vepp += vareij[i][j][(int)age];
12874: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12875: for(j=1;j <=nlstate;j++){
12876: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12877: }
12878: fprintf(ficrest,"\n");
12879: }
1.208 brouard 12880: } /* End vpopbased */
1.269 brouard 12881: free_vector(epj,1,nlstate+1);
1.208 brouard 12882: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12883: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12884: printf("done selection\n");fflush(stdout);
12885: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12886:
1.235 brouard 12887: } /* End k selection */
1.227 brouard 12888:
12889: printf("done State-specific expectancies\n");fflush(stdout);
12890: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12891:
1.288 brouard 12892: /* variance-covariance of forward period prevalence*/
1.269 brouard 12893: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12894:
1.227 brouard 12895:
1.290 brouard 12896: free_vector(weight,firstobs,lastobs);
1.227 brouard 12897: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12898: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12899: free_matrix(anint,1,maxwav,firstobs,lastobs);
12900: free_matrix(mint,1,maxwav,firstobs,lastobs);
12901: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12902: free_ivector(tab,1,NCOVMAX);
12903: fclose(ficresstdeij);
12904: fclose(ficrescveij);
12905: fclose(ficresvij);
12906: fclose(ficrest);
12907: fclose(ficpar);
12908:
12909:
1.126 brouard 12910: /*---------- End : free ----------------*/
1.219 brouard 12911: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12912: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12913: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12914: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12915: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12916: } /* mle==-3 arrives here for freeing */
1.227 brouard 12917: /* endfree:*/
12918: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12919: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12920: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12921: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12922: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12923: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12924: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12925: free_matrix(matcov,1,npar,1,npar);
12926: free_matrix(hess,1,npar,1,npar);
12927: /*free_vector(delti,1,npar);*/
12928: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12929: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12930: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12931: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12932:
12933: free_ivector(ncodemax,1,NCOVMAX);
12934: free_ivector(ncodemaxwundef,1,NCOVMAX);
12935: free_ivector(Dummy,-1,NCOVMAX);
12936: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12937: free_ivector(DummyV,1,NCOVMAX);
12938: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12939: free_ivector(Typevar,-1,NCOVMAX);
12940: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12941: free_ivector(TvarsQ,1,NCOVMAX);
12942: free_ivector(TvarsQind,1,NCOVMAX);
12943: free_ivector(TvarsD,1,NCOVMAX);
12944: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12945: free_ivector(TvarFD,1,NCOVMAX);
12946: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12947: free_ivector(TvarF,1,NCOVMAX);
12948: free_ivector(TvarFind,1,NCOVMAX);
12949: free_ivector(TvarV,1,NCOVMAX);
12950: free_ivector(TvarVind,1,NCOVMAX);
12951: free_ivector(TvarA,1,NCOVMAX);
12952: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12953: free_ivector(TvarFQ,1,NCOVMAX);
12954: free_ivector(TvarFQind,1,NCOVMAX);
12955: free_ivector(TvarVD,1,NCOVMAX);
12956: free_ivector(TvarVDind,1,NCOVMAX);
12957: free_ivector(TvarVQ,1,NCOVMAX);
12958: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12959: free_ivector(Tvarsel,1,NCOVMAX);
12960: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12961: free_ivector(Tposprod,1,NCOVMAX);
12962: free_ivector(Tprod,1,NCOVMAX);
12963: free_ivector(Tvaraff,1,NCOVMAX);
12964: free_ivector(invalidvarcomb,1,ncovcombmax);
12965: free_ivector(Tage,1,NCOVMAX);
12966: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12967: free_ivector(TmodelInvind,1,NCOVMAX);
12968: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12969:
12970: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12971: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12972: fflush(fichtm);
12973: fflush(ficgp);
12974:
1.227 brouard 12975:
1.126 brouard 12976: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12977: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12978: 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 12979: }else{
12980: printf("End of Imach\n");
12981: fprintf(ficlog,"End of Imach\n");
12982: }
12983: printf("See log file on %s\n",filelog);
12984: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12985: /*(void) gettimeofday(&end_time,&tzp);*/
12986: rend_time = time(NULL);
12987: end_time = *localtime(&rend_time);
12988: /* tml = *localtime(&end_time.tm_sec); */
12989: strcpy(strtend,asctime(&end_time));
1.126 brouard 12990: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12991: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12992: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12993:
1.157 brouard 12994: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12995: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12996: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12997: /* printf("Total time was %d uSec.\n", total_usecs);*/
12998: /* if(fileappend(fichtm,optionfilehtm)){ */
12999: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13000: fclose(fichtm);
13001: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13002: fclose(fichtmcov);
13003: fclose(ficgp);
13004: fclose(ficlog);
13005: /*------ End -----------*/
1.227 brouard 13006:
1.281 brouard 13007:
13008: /* Executes gnuplot */
1.227 brouard 13009:
13010: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13011: #ifdef WIN32
1.227 brouard 13012: if (_chdir(pathcd) != 0)
13013: printf("Can't move to directory %s!\n",path);
13014: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13015: #else
1.227 brouard 13016: if(chdir(pathcd) != 0)
13017: printf("Can't move to directory %s!\n", path);
13018: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13019: #endif
1.126 brouard 13020: printf("Current directory %s!\n",pathcd);
13021: /*strcat(plotcmd,CHARSEPARATOR);*/
13022: sprintf(plotcmd,"gnuplot");
1.157 brouard 13023: #ifdef _WIN32
1.126 brouard 13024: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13025: #endif
13026: if(!stat(plotcmd,&info)){
1.158 brouard 13027: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13028: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13029: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13030: }else
13031: strcpy(pplotcmd,plotcmd);
1.157 brouard 13032: #ifdef __unix
1.126 brouard 13033: strcpy(plotcmd,GNUPLOTPROGRAM);
13034: if(!stat(plotcmd,&info)){
1.158 brouard 13035: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13036: }else
13037: strcpy(pplotcmd,plotcmd);
13038: #endif
13039: }else
13040: strcpy(pplotcmd,plotcmd);
13041:
13042: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13043: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13044: strcpy(pplotcmd,plotcmd);
1.227 brouard 13045:
1.126 brouard 13046: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13047: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13048: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13049: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13050: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13051: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13052: strcpy(plotcmd,pplotcmd);
13053: }
1.126 brouard 13054: }
1.158 brouard 13055: printf(" Successful, please wait...");
1.126 brouard 13056: while (z[0] != 'q') {
13057: /* chdir(path); */
1.154 brouard 13058: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13059: scanf("%s",z);
13060: /* if (z[0] == 'c') system("./imach"); */
13061: if (z[0] == 'e') {
1.158 brouard 13062: #ifdef __APPLE__
1.152 brouard 13063: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13064: #elif __linux
13065: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13066: #else
1.152 brouard 13067: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13068: #endif
13069: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13070: system(pplotcmd);
1.126 brouard 13071: }
13072: else if (z[0] == 'g') system(plotcmd);
13073: else if (z[0] == 'q') exit(0);
13074: }
1.227 brouard 13075: end:
1.126 brouard 13076: while (z[0] != 'q') {
1.195 brouard 13077: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13078: scanf("%s",z);
13079: }
1.283 brouard 13080: printf("End\n");
1.282 brouard 13081: exit(0);
1.126 brouard 13082: }
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