Annotation of imach/src/imach.c, revision 1.301
1.301 ! brouard 1: /* $Id: imach.c,v 1.300 2019/05/22 19:09:45 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.301 ! brouard 4: Revision 1.300 2019/05/22 19:09:45 brouard
! 5: Summary: version 0.99r19 of May 2019
! 6:
1.300 brouard 7: Revision 1.299 2019/05/22 18:37:08 brouard
8: Summary: Cleaned 0.99r19
9:
1.299 brouard 10: Revision 1.298 2019/05/22 18:19:56 brouard
11: *** empty log message ***
12:
1.298 brouard 13: Revision 1.297 2019/05/22 17:56:10 brouard
14: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
15:
1.297 brouard 16: Revision 1.296 2019/05/20 13:03:18 brouard
17: Summary: Projection syntax simplified
18:
19:
20: We can now start projections, forward or backward, from the mean date
21: of inteviews up to or down to a number of years of projection:
22: prevforecast=1 yearsfproj=15.3 mobil_average=0
23: or
24: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
25: or
26: prevbackcast=1 yearsbproj=12.3 mobil_average=1
27: or
28: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
29:
1.296 brouard 30: Revision 1.295 2019/05/18 09:52:50 brouard
31: Summary: doxygen tex bug
32:
1.295 brouard 33: Revision 1.294 2019/05/16 14:54:33 brouard
34: Summary: There was some wrong lines added
35:
1.294 brouard 36: Revision 1.293 2019/05/09 15:17:34 brouard
37: *** empty log message ***
38:
1.293 brouard 39: Revision 1.292 2019/05/09 14:17:20 brouard
40: Summary: Some updates
41:
1.292 brouard 42: Revision 1.291 2019/05/09 13:44:18 brouard
43: Summary: Before ncovmax
44:
1.291 brouard 45: Revision 1.290 2019/05/09 13:39:37 brouard
46: Summary: 0.99r18 unlimited number of individuals
47:
48: 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.
49:
1.290 brouard 50: Revision 1.289 2018/12/13 09:16:26 brouard
51: Summary: Bug for young ages (<-30) will be in r17
52:
1.289 brouard 53: Revision 1.288 2018/05/02 20:58:27 brouard
54: Summary: Some bugs fixed
55:
1.288 brouard 56: Revision 1.287 2018/05/01 17:57:25 brouard
57: Summary: Bug fixed by providing frequencies only for non missing covariates
58:
1.287 brouard 59: Revision 1.286 2018/04/27 14:27:04 brouard
60: Summary: some minor bugs
61:
1.286 brouard 62: Revision 1.285 2018/04/21 21:02:16 brouard
63: Summary: Some bugs fixed, valgrind tested
64:
1.285 brouard 65: Revision 1.284 2018/04/20 05:22:13 brouard
66: Summary: Computing mean and stdeviation of fixed quantitative variables
67:
1.284 brouard 68: Revision 1.283 2018/04/19 14:49:16 brouard
69: Summary: Some minor bugs fixed
70:
1.283 brouard 71: Revision 1.282 2018/02/27 22:50:02 brouard
72: *** empty log message ***
73:
1.282 brouard 74: Revision 1.281 2018/02/27 19:25:23 brouard
75: Summary: Adding second argument for quitting
76:
1.281 brouard 77: Revision 1.280 2018/02/21 07:58:13 brouard
78: Summary: 0.99r15
79:
80: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
81:
1.280 brouard 82: Revision 1.279 2017/07/20 13:35:01 brouard
83: Summary: temporary working
84:
1.279 brouard 85: Revision 1.278 2017/07/19 14:09:02 brouard
86: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
87:
1.278 brouard 88: Revision 1.277 2017/07/17 08:53:49 brouard
89: Summary: BOM files can be read now
90:
1.277 brouard 91: Revision 1.276 2017/06/30 15:48:31 brouard
92: Summary: Graphs improvements
93:
1.276 brouard 94: Revision 1.275 2017/06/30 13:39:33 brouard
95: Summary: Saito's color
96:
1.275 brouard 97: Revision 1.274 2017/06/29 09:47:08 brouard
98: Summary: Version 0.99r14
99:
1.274 brouard 100: Revision 1.273 2017/06/27 11:06:02 brouard
101: Summary: More documentation on projections
102:
1.273 brouard 103: Revision 1.272 2017/06/27 10:22:40 brouard
104: Summary: Color of backprojection changed from 6 to 5(yellow)
105:
1.272 brouard 106: Revision 1.271 2017/06/27 10:17:50 brouard
107: Summary: Some bug with rint
108:
1.271 brouard 109: Revision 1.270 2017/05/24 05:45:29 brouard
110: *** empty log message ***
111:
1.270 brouard 112: Revision 1.269 2017/05/23 08:39:25 brouard
113: Summary: Code into subroutine, cleanings
114:
1.269 brouard 115: Revision 1.268 2017/05/18 20:09:32 brouard
116: Summary: backprojection and confidence intervals of backprevalence
117:
1.268 brouard 118: Revision 1.267 2017/05/13 10:25:05 brouard
119: Summary: temporary save for backprojection
120:
1.267 brouard 121: Revision 1.266 2017/05/13 07:26:12 brouard
122: Summary: Version 0.99r13 (improvements and bugs fixed)
123:
1.266 brouard 124: Revision 1.265 2017/04/26 16:22:11 brouard
125: Summary: imach 0.99r13 Some bugs fixed
126:
1.265 brouard 127: Revision 1.264 2017/04/26 06:01:29 brouard
128: Summary: Labels in graphs
129:
1.264 brouard 130: Revision 1.263 2017/04/24 15:23:15 brouard
131: Summary: to save
132:
1.263 brouard 133: Revision 1.262 2017/04/18 16:48:12 brouard
134: *** empty log message ***
135:
1.262 brouard 136: Revision 1.261 2017/04/05 10:14:09 brouard
137: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
138:
1.261 brouard 139: Revision 1.260 2017/04/04 17:46:59 brouard
140: Summary: Gnuplot indexations fixed (humm)
141:
1.260 brouard 142: Revision 1.259 2017/04/04 13:01:16 brouard
143: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
144:
1.259 brouard 145: Revision 1.258 2017/04/03 10:17:47 brouard
146: Summary: Version 0.99r12
147:
148: Some cleanings, conformed with updated documentation.
149:
1.258 brouard 150: Revision 1.257 2017/03/29 16:53:30 brouard
151: Summary: Temp
152:
1.257 brouard 153: Revision 1.256 2017/03/27 05:50:23 brouard
154: Summary: Temporary
155:
1.256 brouard 156: Revision 1.255 2017/03/08 16:02:28 brouard
157: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
158:
1.255 brouard 159: Revision 1.254 2017/03/08 07:13:00 brouard
160: Summary: Fixing data parameter line
161:
1.254 brouard 162: Revision 1.253 2016/12/15 11:59:41 brouard
163: Summary: 0.99 in progress
164:
1.253 brouard 165: Revision 1.252 2016/09/15 21:15:37 brouard
166: *** empty log message ***
167:
1.252 brouard 168: Revision 1.251 2016/09/15 15:01:13 brouard
169: Summary: not working
170:
1.251 brouard 171: Revision 1.250 2016/09/08 16:07:27 brouard
172: Summary: continue
173:
1.250 brouard 174: Revision 1.249 2016/09/07 17:14:18 brouard
175: Summary: Starting values from frequencies
176:
1.249 brouard 177: Revision 1.248 2016/09/07 14:10:18 brouard
178: *** empty log message ***
179:
1.248 brouard 180: Revision 1.247 2016/09/02 11:11:21 brouard
181: *** empty log message ***
182:
1.247 brouard 183: Revision 1.246 2016/09/02 08:49:22 brouard
184: *** empty log message ***
185:
1.246 brouard 186: Revision 1.245 2016/09/02 07:25:01 brouard
187: *** empty log message ***
188:
1.245 brouard 189: Revision 1.244 2016/09/02 07:17:34 brouard
190: *** empty log message ***
191:
1.244 brouard 192: Revision 1.243 2016/09/02 06:45:35 brouard
193: *** empty log message ***
194:
1.243 brouard 195: Revision 1.242 2016/08/30 15:01:20 brouard
196: Summary: Fixing a lots
197:
1.242 brouard 198: Revision 1.241 2016/08/29 17:17:25 brouard
199: Summary: gnuplot problem in Back projection to fix
200:
1.241 brouard 201: Revision 1.240 2016/08/29 07:53:18 brouard
202: Summary: Better
203:
1.240 brouard 204: Revision 1.239 2016/08/26 15:51:03 brouard
205: Summary: Improvement in Powell output in order to copy and paste
206:
207: Author:
208:
1.239 brouard 209: Revision 1.238 2016/08/26 14:23:35 brouard
210: Summary: Starting tests of 0.99
211:
1.238 brouard 212: Revision 1.237 2016/08/26 09:20:19 brouard
213: Summary: to valgrind
214:
1.237 brouard 215: Revision 1.236 2016/08/25 10:50:18 brouard
216: *** empty log message ***
217:
1.236 brouard 218: Revision 1.235 2016/08/25 06:59:23 brouard
219: *** empty log message ***
220:
1.235 brouard 221: Revision 1.234 2016/08/23 16:51:20 brouard
222: *** empty log message ***
223:
1.234 brouard 224: Revision 1.233 2016/08/23 07:40:50 brouard
225: Summary: not working
226:
1.233 brouard 227: Revision 1.232 2016/08/22 14:20:21 brouard
228: Summary: not working
229:
1.232 brouard 230: Revision 1.231 2016/08/22 07:17:15 brouard
231: Summary: not working
232:
1.231 brouard 233: Revision 1.230 2016/08/22 06:55:53 brouard
234: Summary: Not working
235:
1.230 brouard 236: Revision 1.229 2016/07/23 09:45:53 brouard
237: Summary: Completing for func too
238:
1.229 brouard 239: Revision 1.228 2016/07/22 17:45:30 brouard
240: Summary: Fixing some arrays, still debugging
241:
1.227 brouard 242: Revision 1.226 2016/07/12 18:42:34 brouard
243: Summary: temp
244:
1.226 brouard 245: Revision 1.225 2016/07/12 08:40:03 brouard
246: Summary: saving but not running
247:
1.225 brouard 248: Revision 1.224 2016/07/01 13:16:01 brouard
249: Summary: Fixes
250:
1.224 brouard 251: Revision 1.223 2016/02/19 09:23:35 brouard
252: Summary: temporary
253:
1.223 brouard 254: Revision 1.222 2016/02/17 08:14:50 brouard
255: Summary: Probably last 0.98 stable version 0.98r6
256:
1.222 brouard 257: Revision 1.221 2016/02/15 23:35:36 brouard
258: Summary: minor bug
259:
1.220 brouard 260: Revision 1.219 2016/02/15 00:48:12 brouard
261: *** empty log message ***
262:
1.219 brouard 263: Revision 1.218 2016/02/12 11:29:23 brouard
264: Summary: 0.99 Back projections
265:
1.218 brouard 266: Revision 1.217 2015/12/23 17:18:31 brouard
267: Summary: Experimental backcast
268:
1.217 brouard 269: Revision 1.216 2015/12/18 17:32:11 brouard
270: Summary: 0.98r4 Warning and status=-2
271:
272: Version 0.98r4 is now:
273: - displaying an error when status is -1, date of interview unknown and date of death known;
274: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
275: Older changes concerning s=-2, dating from 2005 have been supersed.
276:
1.216 brouard 277: Revision 1.215 2015/12/16 08:52:24 brouard
278: Summary: 0.98r4 working
279:
1.215 brouard 280: Revision 1.214 2015/12/16 06:57:54 brouard
281: Summary: temporary not working
282:
1.214 brouard 283: Revision 1.213 2015/12/11 18:22:17 brouard
284: Summary: 0.98r4
285:
1.213 brouard 286: Revision 1.212 2015/11/21 12:47:24 brouard
287: Summary: minor typo
288:
1.212 brouard 289: Revision 1.211 2015/11/21 12:41:11 brouard
290: Summary: 0.98r3 with some graph of projected cross-sectional
291:
292: Author: Nicolas Brouard
293:
1.211 brouard 294: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 295: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 296: Summary: Adding ftolpl parameter
297: Author: N Brouard
298:
299: We had difficulties to get smoothed confidence intervals. It was due
300: to the period prevalence which wasn't computed accurately. The inner
301: parameter ftolpl is now an outer parameter of the .imach parameter
302: file after estepm. If ftolpl is small 1.e-4 and estepm too,
303: computation are long.
304:
1.209 brouard 305: Revision 1.208 2015/11/17 14:31:57 brouard
306: Summary: temporary
307:
1.208 brouard 308: Revision 1.207 2015/10/27 17:36:57 brouard
309: *** empty log message ***
310:
1.207 brouard 311: Revision 1.206 2015/10/24 07:14:11 brouard
312: *** empty log message ***
313:
1.206 brouard 314: Revision 1.205 2015/10/23 15:50:53 brouard
315: Summary: 0.98r3 some clarification for graphs on likelihood contributions
316:
1.205 brouard 317: Revision 1.204 2015/10/01 16:20:26 brouard
318: Summary: Some new graphs of contribution to likelihood
319:
1.204 brouard 320: Revision 1.203 2015/09/30 17:45:14 brouard
321: Summary: looking at better estimation of the hessian
322:
323: Also a better criteria for convergence to the period prevalence And
324: therefore adding the number of years needed to converge. (The
325: prevalence in any alive state shold sum to one
326:
1.203 brouard 327: Revision 1.202 2015/09/22 19:45:16 brouard
328: Summary: Adding some overall graph on contribution to likelihood. Might change
329:
1.202 brouard 330: Revision 1.201 2015/09/15 17:34:58 brouard
331: Summary: 0.98r0
332:
333: - Some new graphs like suvival functions
334: - Some bugs fixed like model=1+age+V2.
335:
1.201 brouard 336: Revision 1.200 2015/09/09 16:53:55 brouard
337: Summary: Big bug thanks to Flavia
338:
339: Even model=1+age+V2. did not work anymore
340:
1.200 brouard 341: Revision 1.199 2015/09/07 14:09:23 brouard
342: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
343:
1.199 brouard 344: Revision 1.198 2015/09/03 07:14:39 brouard
345: Summary: 0.98q5 Flavia
346:
1.198 brouard 347: Revision 1.197 2015/09/01 18:24:39 brouard
348: *** empty log message ***
349:
1.197 brouard 350: Revision 1.196 2015/08/18 23:17:52 brouard
351: Summary: 0.98q5
352:
1.196 brouard 353: Revision 1.195 2015/08/18 16:28:39 brouard
354: Summary: Adding a hack for testing purpose
355:
356: After reading the title, ftol and model lines, if the comment line has
357: a q, starting with #q, the answer at the end of the run is quit. It
358: permits to run test files in batch with ctest. The former workaround was
359: $ echo q | imach foo.imach
360:
1.195 brouard 361: Revision 1.194 2015/08/18 13:32:00 brouard
362: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
363:
1.194 brouard 364: Revision 1.193 2015/08/04 07:17:42 brouard
365: Summary: 0.98q4
366:
1.193 brouard 367: Revision 1.192 2015/07/16 16:49:02 brouard
368: Summary: Fixing some outputs
369:
1.192 brouard 370: Revision 1.191 2015/07/14 10:00:33 brouard
371: Summary: Some fixes
372:
1.191 brouard 373: Revision 1.190 2015/05/05 08:51:13 brouard
374: Summary: Adding digits in output parameters (7 digits instead of 6)
375:
376: Fix 1+age+.
377:
1.190 brouard 378: Revision 1.189 2015/04/30 14:45:16 brouard
379: Summary: 0.98q2
380:
1.189 brouard 381: Revision 1.188 2015/04/30 08:27:53 brouard
382: *** empty log message ***
383:
1.188 brouard 384: Revision 1.187 2015/04/29 09:11:15 brouard
385: *** empty log message ***
386:
1.187 brouard 387: Revision 1.186 2015/04/23 12:01:52 brouard
388: Summary: V1*age is working now, version 0.98q1
389:
390: Some codes had been disabled in order to simplify and Vn*age was
391: working in the optimization phase, ie, giving correct MLE parameters,
392: but, as usual, outputs were not correct and program core dumped.
393:
1.186 brouard 394: Revision 1.185 2015/03/11 13:26:42 brouard
395: Summary: Inclusion of compile and links command line for Intel Compiler
396:
1.185 brouard 397: Revision 1.184 2015/03/11 11:52:39 brouard
398: Summary: Back from Windows 8. Intel Compiler
399:
1.184 brouard 400: Revision 1.183 2015/03/10 20:34:32 brouard
401: Summary: 0.98q0, trying with directest, mnbrak fixed
402:
403: We use directest instead of original Powell test; probably no
404: incidence on the results, but better justifications;
405: We fixed Numerical Recipes mnbrak routine which was wrong and gave
406: wrong results.
407:
1.183 brouard 408: Revision 1.182 2015/02/12 08:19:57 brouard
409: Summary: Trying to keep directest which seems simpler and more general
410: Author: Nicolas Brouard
411:
1.182 brouard 412: Revision 1.181 2015/02/11 23:22:24 brouard
413: Summary: Comments on Powell added
414:
415: Author:
416:
1.181 brouard 417: Revision 1.180 2015/02/11 17:33:45 brouard
418: Summary: Finishing move from main to function (hpijx and prevalence_limit)
419:
1.180 brouard 420: Revision 1.179 2015/01/04 09:57:06 brouard
421: Summary: back to OS/X
422:
1.179 brouard 423: Revision 1.178 2015/01/04 09:35:48 brouard
424: *** empty log message ***
425:
1.178 brouard 426: Revision 1.177 2015/01/03 18:40:56 brouard
427: Summary: Still testing ilc32 on OSX
428:
1.177 brouard 429: Revision 1.176 2015/01/03 16:45:04 brouard
430: *** empty log message ***
431:
1.176 brouard 432: Revision 1.175 2015/01/03 16:33:42 brouard
433: *** empty log message ***
434:
1.175 brouard 435: Revision 1.174 2015/01/03 16:15:49 brouard
436: Summary: Still in cross-compilation
437:
1.174 brouard 438: Revision 1.173 2015/01/03 12:06:26 brouard
439: Summary: trying to detect cross-compilation
440:
1.173 brouard 441: Revision 1.172 2014/12/27 12:07:47 brouard
442: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
443:
1.172 brouard 444: Revision 1.171 2014/12/23 13:26:59 brouard
445: Summary: Back from Visual C
446:
447: Still problem with utsname.h on Windows
448:
1.171 brouard 449: Revision 1.170 2014/12/23 11:17:12 brouard
450: Summary: Cleaning some \%% back to %%
451:
452: The escape was mandatory for a specific compiler (which one?), but too many warnings.
453:
1.170 brouard 454: Revision 1.169 2014/12/22 23:08:31 brouard
455: Summary: 0.98p
456:
457: Outputs some informations on compiler used, OS etc. Testing on different platforms.
458:
1.169 brouard 459: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 460: Summary: update
1.169 brouard 461:
1.168 brouard 462: Revision 1.167 2014/12/22 13:50:56 brouard
463: Summary: Testing uname and compiler version and if compiled 32 or 64
464:
465: Testing on Linux 64
466:
1.167 brouard 467: Revision 1.166 2014/12/22 11:40:47 brouard
468: *** empty log message ***
469:
1.166 brouard 470: Revision 1.165 2014/12/16 11:20:36 brouard
471: Summary: After compiling on Visual C
472:
473: * imach.c (Module): Merging 1.61 to 1.162
474:
1.165 brouard 475: Revision 1.164 2014/12/16 10:52:11 brouard
476: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
477:
478: * imach.c (Module): Merging 1.61 to 1.162
479:
1.164 brouard 480: Revision 1.163 2014/12/16 10:30:11 brouard
481: * imach.c (Module): Merging 1.61 to 1.162
482:
1.163 brouard 483: Revision 1.162 2014/09/25 11:43:39 brouard
484: Summary: temporary backup 0.99!
485:
1.162 brouard 486: Revision 1.1 2014/09/16 11:06:58 brouard
487: Summary: With some code (wrong) for nlopt
488:
489: Author:
490:
491: Revision 1.161 2014/09/15 20:41:41 brouard
492: Summary: Problem with macro SQR on Intel compiler
493:
1.161 brouard 494: Revision 1.160 2014/09/02 09:24:05 brouard
495: *** empty log message ***
496:
1.160 brouard 497: Revision 1.159 2014/09/01 10:34:10 brouard
498: Summary: WIN32
499: Author: Brouard
500:
1.159 brouard 501: Revision 1.158 2014/08/27 17:11:51 brouard
502: *** empty log message ***
503:
1.158 brouard 504: Revision 1.157 2014/08/27 16:26:55 brouard
505: Summary: Preparing windows Visual studio version
506: Author: Brouard
507:
508: In order to compile on Visual studio, time.h is now correct and time_t
509: and tm struct should be used. difftime should be used but sometimes I
510: just make the differences in raw time format (time(&now).
511: Trying to suppress #ifdef LINUX
512: Add xdg-open for __linux in order to open default browser.
513:
1.157 brouard 514: Revision 1.156 2014/08/25 20:10:10 brouard
515: *** empty log message ***
516:
1.156 brouard 517: Revision 1.155 2014/08/25 18:32:34 brouard
518: Summary: New compile, minor changes
519: Author: Brouard
520:
1.155 brouard 521: Revision 1.154 2014/06/20 17:32:08 brouard
522: Summary: Outputs now all graphs of convergence to period prevalence
523:
1.154 brouard 524: Revision 1.153 2014/06/20 16:45:46 brouard
525: Summary: If 3 live state, convergence to period prevalence on same graph
526: Author: Brouard
527:
1.153 brouard 528: Revision 1.152 2014/06/18 17:54:09 brouard
529: Summary: open browser, use gnuplot on same dir than imach if not found in the path
530:
1.152 brouard 531: Revision 1.151 2014/06/18 16:43:30 brouard
532: *** empty log message ***
533:
1.151 brouard 534: Revision 1.150 2014/06/18 16:42:35 brouard
535: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
536: Author: brouard
537:
1.150 brouard 538: Revision 1.149 2014/06/18 15:51:14 brouard
539: Summary: Some fixes in parameter files errors
540: Author: Nicolas Brouard
541:
1.149 brouard 542: Revision 1.148 2014/06/17 17:38:48 brouard
543: Summary: Nothing new
544: Author: Brouard
545:
546: Just a new packaging for OS/X version 0.98nS
547:
1.148 brouard 548: Revision 1.147 2014/06/16 10:33:11 brouard
549: *** empty log message ***
550:
1.147 brouard 551: Revision 1.146 2014/06/16 10:20:28 brouard
552: Summary: Merge
553: Author: Brouard
554:
555: Merge, before building revised version.
556:
1.146 brouard 557: Revision 1.145 2014/06/10 21:23:15 brouard
558: Summary: Debugging with valgrind
559: Author: Nicolas Brouard
560:
561: Lot of changes in order to output the results with some covariates
562: After the Edimburgh REVES conference 2014, it seems mandatory to
563: improve the code.
564: No more memory valgrind error but a lot has to be done in order to
565: continue the work of splitting the code into subroutines.
566: Also, decodemodel has been improved. Tricode is still not
567: optimal. nbcode should be improved. Documentation has been added in
568: the source code.
569:
1.144 brouard 570: Revision 1.143 2014/01/26 09:45:38 brouard
571: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
572:
573: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
574: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
575:
1.143 brouard 576: Revision 1.142 2014/01/26 03:57:36 brouard
577: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
578:
579: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
580:
1.142 brouard 581: Revision 1.141 2014/01/26 02:42:01 brouard
582: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
583:
1.141 brouard 584: Revision 1.140 2011/09/02 10:37:54 brouard
585: Summary: times.h is ok with mingw32 now.
586:
1.140 brouard 587: Revision 1.139 2010/06/14 07:50:17 brouard
588: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
589: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
590:
1.139 brouard 591: Revision 1.138 2010/04/30 18:19:40 brouard
592: *** empty log message ***
593:
1.138 brouard 594: Revision 1.137 2010/04/29 18:11:38 brouard
595: (Module): Checking covariates for more complex models
596: than V1+V2. A lot of change to be done. Unstable.
597:
1.137 brouard 598: Revision 1.136 2010/04/26 20:30:53 brouard
599: (Module): merging some libgsl code. Fixing computation
600: of likelione (using inter/intrapolation if mle = 0) in order to
601: get same likelihood as if mle=1.
602: Some cleaning of code and comments added.
603:
1.136 brouard 604: Revision 1.135 2009/10/29 15:33:14 brouard
605: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
606:
1.135 brouard 607: Revision 1.134 2009/10/29 13:18:53 brouard
608: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
609:
1.134 brouard 610: Revision 1.133 2009/07/06 10:21:25 brouard
611: just nforces
612:
1.133 brouard 613: Revision 1.132 2009/07/06 08:22:05 brouard
614: Many tings
615:
1.132 brouard 616: Revision 1.131 2009/06/20 16:22:47 brouard
617: Some dimensions resccaled
618:
1.131 brouard 619: Revision 1.130 2009/05/26 06:44:34 brouard
620: (Module): Max Covariate is now set to 20 instead of 8. A
621: lot of cleaning with variables initialized to 0. Trying to make
622: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
623:
1.130 brouard 624: Revision 1.129 2007/08/31 13:49:27 lievre
625: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
626:
1.129 lievre 627: Revision 1.128 2006/06/30 13:02:05 brouard
628: (Module): Clarifications on computing e.j
629:
1.128 brouard 630: Revision 1.127 2006/04/28 18:11:50 brouard
631: (Module): Yes the sum of survivors was wrong since
632: imach-114 because nhstepm was no more computed in the age
633: loop. Now we define nhstepma in the age loop.
634: (Module): In order to speed up (in case of numerous covariates) we
635: compute health expectancies (without variances) in a first step
636: and then all the health expectancies with variances or standard
637: deviation (needs data from the Hessian matrices) which slows the
638: computation.
639: In the future we should be able to stop the program is only health
640: expectancies and graph are needed without standard deviations.
641:
1.127 brouard 642: Revision 1.126 2006/04/28 17:23:28 brouard
643: (Module): Yes the sum of survivors was wrong since
644: imach-114 because nhstepm was no more computed in the age
645: loop. Now we define nhstepma in the age loop.
646: Version 0.98h
647:
1.126 brouard 648: Revision 1.125 2006/04/04 15:20:31 lievre
649: Errors in calculation of health expectancies. Age was not initialized.
650: Forecasting file added.
651:
652: Revision 1.124 2006/03/22 17:13:53 lievre
653: Parameters are printed with %lf instead of %f (more numbers after the comma).
654: The log-likelihood is printed in the log file
655:
656: Revision 1.123 2006/03/20 10:52:43 brouard
657: * imach.c (Module): <title> changed, corresponds to .htm file
658: name. <head> headers where missing.
659:
660: * imach.c (Module): Weights can have a decimal point as for
661: English (a comma might work with a correct LC_NUMERIC environment,
662: otherwise the weight is truncated).
663: Modification of warning when the covariates values are not 0 or
664: 1.
665: Version 0.98g
666:
667: Revision 1.122 2006/03/20 09:45:41 brouard
668: (Module): Weights can have a decimal point as for
669: English (a comma might work with a correct LC_NUMERIC environment,
670: otherwise the weight is truncated).
671: Modification of warning when the covariates values are not 0 or
672: 1.
673: Version 0.98g
674:
675: Revision 1.121 2006/03/16 17:45:01 lievre
676: * imach.c (Module): Comments concerning covariates added
677:
678: * imach.c (Module): refinements in the computation of lli if
679: status=-2 in order to have more reliable computation if stepm is
680: not 1 month. Version 0.98f
681:
682: Revision 1.120 2006/03/16 15:10:38 lievre
683: (Module): refinements in the computation of lli if
684: status=-2 in order to have more reliable computation if stepm is
685: not 1 month. Version 0.98f
686:
687: Revision 1.119 2006/03/15 17:42:26 brouard
688: (Module): Bug if status = -2, the loglikelihood was
689: computed as likelihood omitting the logarithm. Version O.98e
690:
691: Revision 1.118 2006/03/14 18:20:07 brouard
692: (Module): varevsij Comments added explaining the second
693: table of variances if popbased=1 .
694: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
695: (Module): Function pstamp added
696: (Module): Version 0.98d
697:
698: Revision 1.117 2006/03/14 17:16:22 brouard
699: (Module): varevsij Comments added explaining the second
700: table of variances if popbased=1 .
701: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
702: (Module): Function pstamp added
703: (Module): Version 0.98d
704:
705: Revision 1.116 2006/03/06 10:29:27 brouard
706: (Module): Variance-covariance wrong links and
707: varian-covariance of ej. is needed (Saito).
708:
709: Revision 1.115 2006/02/27 12:17:45 brouard
710: (Module): One freematrix added in mlikeli! 0.98c
711:
712: Revision 1.114 2006/02/26 12:57:58 brouard
713: (Module): Some improvements in processing parameter
714: filename with strsep.
715:
716: Revision 1.113 2006/02/24 14:20:24 brouard
717: (Module): Memory leaks checks with valgrind and:
718: datafile was not closed, some imatrix were not freed and on matrix
719: allocation too.
720:
721: Revision 1.112 2006/01/30 09:55:26 brouard
722: (Module): Back to gnuplot.exe instead of wgnuplot.exe
723:
724: Revision 1.111 2006/01/25 20:38:18 brouard
725: (Module): Lots of cleaning and bugs added (Gompertz)
726: (Module): Comments can be added in data file. Missing date values
727: can be a simple dot '.'.
728:
729: Revision 1.110 2006/01/25 00:51:50 brouard
730: (Module): Lots of cleaning and bugs added (Gompertz)
731:
732: Revision 1.109 2006/01/24 19:37:15 brouard
733: (Module): Comments (lines starting with a #) are allowed in data.
734:
735: Revision 1.108 2006/01/19 18:05:42 lievre
736: Gnuplot problem appeared...
737: To be fixed
738:
739: Revision 1.107 2006/01/19 16:20:37 brouard
740: Test existence of gnuplot in imach path
741:
742: Revision 1.106 2006/01/19 13:24:36 brouard
743: Some cleaning and links added in html output
744:
745: Revision 1.105 2006/01/05 20:23:19 lievre
746: *** empty log message ***
747:
748: Revision 1.104 2005/09/30 16:11:43 lievre
749: (Module): sump fixed, loop imx fixed, and simplifications.
750: (Module): If the status is missing at the last wave but we know
751: that the person is alive, then we can code his/her status as -2
752: (instead of missing=-1 in earlier versions) and his/her
753: contributions to the likelihood is 1 - Prob of dying from last
754: health status (= 1-p13= p11+p12 in the easiest case of somebody in
755: the healthy state at last known wave). Version is 0.98
756:
757: Revision 1.103 2005/09/30 15:54:49 lievre
758: (Module): sump fixed, loop imx fixed, and simplifications.
759:
760: Revision 1.102 2004/09/15 17:31:30 brouard
761: Add the possibility to read data file including tab characters.
762:
763: Revision 1.101 2004/09/15 10:38:38 brouard
764: Fix on curr_time
765:
766: Revision 1.100 2004/07/12 18:29:06 brouard
767: Add version for Mac OS X. Just define UNIX in Makefile
768:
769: Revision 1.99 2004/06/05 08:57:40 brouard
770: *** empty log message ***
771:
772: Revision 1.98 2004/05/16 15:05:56 brouard
773: New version 0.97 . First attempt to estimate force of mortality
774: directly from the data i.e. without the need of knowing the health
775: state at each age, but using a Gompertz model: log u =a + b*age .
776: This is the basic analysis of mortality and should be done before any
777: other analysis, in order to test if the mortality estimated from the
778: cross-longitudinal survey is different from the mortality estimated
779: from other sources like vital statistic data.
780:
781: The same imach parameter file can be used but the option for mle should be -3.
782:
1.133 brouard 783: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 784: former routines in order to include the new code within the former code.
785:
786: The output is very simple: only an estimate of the intercept and of
787: the slope with 95% confident intervals.
788:
789: Current limitations:
790: A) Even if you enter covariates, i.e. with the
791: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
792: B) There is no computation of Life Expectancy nor Life Table.
793:
794: Revision 1.97 2004/02/20 13:25:42 lievre
795: Version 0.96d. Population forecasting command line is (temporarily)
796: suppressed.
797:
798: Revision 1.96 2003/07/15 15:38:55 brouard
799: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
800: rewritten within the same printf. Workaround: many printfs.
801:
802: Revision 1.95 2003/07/08 07:54:34 brouard
803: * imach.c (Repository):
804: (Repository): Using imachwizard code to output a more meaningful covariance
805: matrix (cov(a12,c31) instead of numbers.
806:
807: Revision 1.94 2003/06/27 13:00:02 brouard
808: Just cleaning
809:
810: Revision 1.93 2003/06/25 16:33:55 brouard
811: (Module): On windows (cygwin) function asctime_r doesn't
812: exist so I changed back to asctime which exists.
813: (Module): Version 0.96b
814:
815: Revision 1.92 2003/06/25 16:30:45 brouard
816: (Module): On windows (cygwin) function asctime_r doesn't
817: exist so I changed back to asctime which exists.
818:
819: Revision 1.91 2003/06/25 15:30:29 brouard
820: * imach.c (Repository): Duplicated warning errors corrected.
821: (Repository): Elapsed time after each iteration is now output. It
822: helps to forecast when convergence will be reached. Elapsed time
823: is stamped in powell. We created a new html file for the graphs
824: concerning matrix of covariance. It has extension -cov.htm.
825:
826: Revision 1.90 2003/06/24 12:34:15 brouard
827: (Module): Some bugs corrected for windows. Also, when
828: mle=-1 a template is output in file "or"mypar.txt with the design
829: of the covariance matrix to be input.
830:
831: Revision 1.89 2003/06/24 12:30:52 brouard
832: (Module): Some bugs corrected for windows. Also, when
833: mle=-1 a template is output in file "or"mypar.txt with the design
834: of the covariance matrix to be input.
835:
836: Revision 1.88 2003/06/23 17:54:56 brouard
837: * 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.
838:
839: Revision 1.87 2003/06/18 12:26:01 brouard
840: Version 0.96
841:
842: Revision 1.86 2003/06/17 20:04:08 brouard
843: (Module): Change position of html and gnuplot routines and added
844: routine fileappend.
845:
846: Revision 1.85 2003/06/17 13:12:43 brouard
847: * imach.c (Repository): Check when date of death was earlier that
848: current date of interview. It may happen when the death was just
849: prior to the death. In this case, dh was negative and likelihood
850: was wrong (infinity). We still send an "Error" but patch by
851: assuming that the date of death was just one stepm after the
852: interview.
853: (Repository): Because some people have very long ID (first column)
854: we changed int to long in num[] and we added a new lvector for
855: memory allocation. But we also truncated to 8 characters (left
856: truncation)
857: (Repository): No more line truncation errors.
858:
859: Revision 1.84 2003/06/13 21:44:43 brouard
860: * imach.c (Repository): Replace "freqsummary" at a correct
861: place. It differs from routine "prevalence" which may be called
862: many times. Probs is memory consuming and must be used with
863: parcimony.
864: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
865:
866: Revision 1.83 2003/06/10 13:39:11 lievre
867: *** empty log message ***
868:
869: Revision 1.82 2003/06/05 15:57:20 brouard
870: Add log in imach.c and fullversion number is now printed.
871:
872: */
873: /*
874: Interpolated Markov Chain
875:
876: Short summary of the programme:
877:
1.227 brouard 878: This program computes Healthy Life Expectancies or State-specific
879: (if states aren't health statuses) Expectancies from
880: cross-longitudinal data. Cross-longitudinal data consist in:
881:
882: -1- a first survey ("cross") where individuals from different ages
883: are interviewed on their health status or degree of disability (in
884: the case of a health survey which is our main interest)
885:
886: -2- at least a second wave of interviews ("longitudinal") which
887: measure each change (if any) in individual health status. Health
888: expectancies are computed from the time spent in each health state
889: according to a model. More health states you consider, more time is
890: necessary to reach the Maximum Likelihood of the parameters involved
891: in the model. The simplest model is the multinomial logistic model
892: where pij is the probability to be observed in state j at the second
893: wave conditional to be observed in state i at the first
894: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
895: etc , where 'age' is age and 'sex' is a covariate. If you want to
896: have a more complex model than "constant and age", you should modify
897: the program where the markup *Covariates have to be included here
898: again* invites you to do it. More covariates you add, slower the
1.126 brouard 899: convergence.
900:
901: The advantage of this computer programme, compared to a simple
902: multinomial logistic model, is clear when the delay between waves is not
903: identical for each individual. Also, if a individual missed an
904: intermediate interview, the information is lost, but taken into
905: account using an interpolation or extrapolation.
906:
907: hPijx is the probability to be observed in state i at age x+h
908: conditional to the observed state i at age x. The delay 'h' can be
909: split into an exact number (nh*stepm) of unobserved intermediate
910: states. This elementary transition (by month, quarter,
911: semester or year) is modelled as a multinomial logistic. The hPx
912: matrix is simply the matrix product of nh*stepm elementary matrices
913: and the contribution of each individual to the likelihood is simply
914: hPijx.
915:
916: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 917: of the life expectancies. It also computes the period (stable) prevalence.
918:
919: Back prevalence and projections:
1.227 brouard 920:
921: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
922: double agemaxpar, double ftolpl, int *ncvyearp, double
923: dateprev1,double dateprev2, int firstpass, int lastpass, int
924: mobilavproj)
925:
926: Computes the back prevalence limit for any combination of
927: covariate values k at any age between ageminpar and agemaxpar and
928: returns it in **bprlim. In the loops,
929:
930: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
931: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
932:
933: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 934: Computes for any combination of covariates k and any age between bage and fage
935: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
936: oldm=oldms;savm=savms;
1.227 brouard 937:
1.267 brouard 938: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 939: Computes the transition matrix starting at age 'age' over
940: 'nhstepm*hstepm*stepm' months (i.e. until
941: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 942: nhstepm*hstepm matrices.
943:
944: Returns p3mat[i][j][h] after calling
945: p3mat[i][j][h]=matprod2(newm,
946: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
947: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
948: oldm);
1.226 brouard 949:
950: Important routines
951:
952: - func (or funcone), computes logit (pij) distinguishing
953: o fixed variables (single or product dummies or quantitative);
954: o varying variables by:
955: (1) wave (single, product dummies, quantitative),
956: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
957: % fixed dummy (treated) or quantitative (not done because time-consuming);
958: % varying dummy (not done) or quantitative (not done);
959: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
960: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
961: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
962: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
963: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 964:
1.226 brouard 965:
966:
1.133 brouard 967: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
968: Institut national d'études démographiques, Paris.
1.126 brouard 969: This software have been partly granted by Euro-REVES, a concerted action
970: from the European Union.
971: It is copyrighted identically to a GNU software product, ie programme and
972: software can be distributed freely for non commercial use. Latest version
973: can be accessed at http://euroreves.ined.fr/imach .
974:
975: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
976: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
977:
978: **********************************************************************/
979: /*
980: main
981: read parameterfile
982: read datafile
983: concatwav
984: freqsummary
985: if (mle >= 1)
986: mlikeli
987: print results files
988: if mle==1
989: computes hessian
990: read end of parameter file: agemin, agemax, bage, fage, estepm
991: begin-prev-date,...
992: open gnuplot file
993: open html file
1.145 brouard 994: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
995: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
996: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
997: freexexit2 possible for memory heap.
998:
999: h Pij x | pij_nom ficrestpij
1000: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1001: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1002: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1003:
1004: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1005: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1006: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1007: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1008: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1009:
1.126 brouard 1010: forecasting if prevfcast==1 prevforecast call prevalence()
1011: health expectancies
1012: Variance-covariance of DFLE
1013: prevalence()
1014: movingaverage()
1015: varevsij()
1016: if popbased==1 varevsij(,popbased)
1017: total life expectancies
1018: Variance of period (stable) prevalence
1019: end
1020: */
1021:
1.187 brouard 1022: /* #define DEBUG */
1023: /* #define DEBUGBRENT */
1.203 brouard 1024: /* #define DEBUGLINMIN */
1025: /* #define DEBUGHESS */
1026: #define DEBUGHESSIJ
1.224 brouard 1027: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1028: #define POWELL /* Instead of NLOPT */
1.224 brouard 1029: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1030: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1031: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1032:
1033: #include <math.h>
1034: #include <stdio.h>
1035: #include <stdlib.h>
1036: #include <string.h>
1.226 brouard 1037: #include <ctype.h>
1.159 brouard 1038:
1039: #ifdef _WIN32
1040: #include <io.h>
1.172 brouard 1041: #include <windows.h>
1042: #include <tchar.h>
1.159 brouard 1043: #else
1.126 brouard 1044: #include <unistd.h>
1.159 brouard 1045: #endif
1.126 brouard 1046:
1047: #include <limits.h>
1048: #include <sys/types.h>
1.171 brouard 1049:
1050: #if defined(__GNUC__)
1051: #include <sys/utsname.h> /* Doesn't work on Windows */
1052: #endif
1053:
1.126 brouard 1054: #include <sys/stat.h>
1055: #include <errno.h>
1.159 brouard 1056: /* extern int errno; */
1.126 brouard 1057:
1.157 brouard 1058: /* #ifdef LINUX */
1059: /* #include <time.h> */
1060: /* #include "timeval.h" */
1061: /* #else */
1062: /* #include <sys/time.h> */
1063: /* #endif */
1064:
1.126 brouard 1065: #include <time.h>
1066:
1.136 brouard 1067: #ifdef GSL
1068: #include <gsl/gsl_errno.h>
1069: #include <gsl/gsl_multimin.h>
1070: #endif
1071:
1.167 brouard 1072:
1.162 brouard 1073: #ifdef NLOPT
1074: #include <nlopt.h>
1075: typedef struct {
1076: double (* function)(double [] );
1077: } myfunc_data ;
1078: #endif
1079:
1.126 brouard 1080: /* #include <libintl.h> */
1081: /* #define _(String) gettext (String) */
1082:
1.251 brouard 1083: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1084:
1085: #define GNUPLOTPROGRAM "gnuplot"
1086: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1087: #define FILENAMELENGTH 132
1088:
1089: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1090: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1091:
1.144 brouard 1092: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1093: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1094:
1095: #define NINTERVMAX 8
1.144 brouard 1096: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1097: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1098: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1099: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1100: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1101: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1102: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1103: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1104: /* #define AGESUP 130 */
1.288 brouard 1105: /* #define AGESUP 150 */
1106: #define AGESUP 200
1.268 brouard 1107: #define AGEINF 0
1.218 brouard 1108: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1109: #define AGEBASE 40
1.194 brouard 1110: #define AGEOVERFLOW 1.e20
1.164 brouard 1111: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1112: #ifdef _WIN32
1113: #define DIRSEPARATOR '\\'
1114: #define CHARSEPARATOR "\\"
1115: #define ODIRSEPARATOR '/'
1116: #else
1.126 brouard 1117: #define DIRSEPARATOR '/'
1118: #define CHARSEPARATOR "/"
1119: #define ODIRSEPARATOR '\\'
1120: #endif
1121:
1.301 ! brouard 1122: /* $Id: imach.c,v 1.300 2019/05/22 19:09:45 brouard Exp $ */
1.126 brouard 1123: /* $State: Exp $ */
1.196 brouard 1124: #include "version.h"
1125: char version[]=__IMACH_VERSION__;
1.300 brouard 1126: 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";
1.301 ! brouard 1127: char fullversion[]="$Revision: 1.300 $ $Date: 2019/05/22 19:09:45 $";
1.126 brouard 1128: char strstart[80];
1129: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1130: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1131: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1132: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1133: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1134: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1135: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1136: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1137: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1138: int cptcovprodnoage=0; /**< Number of covariate products without age */
1139: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1140: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1141: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1142: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1143: int nsd=0; /**< Total number of single dummy variables (output) */
1144: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1145: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1146: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1147: int ntveff=0; /**< ntveff number of effective time varying variables */
1148: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1149: int cptcov=0; /* Working variable */
1.290 brouard 1150: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1151: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1152: int npar=NPARMAX;
1153: int nlstate=2; /* Number of live states */
1154: int ndeath=1; /* Number of dead states */
1.130 brouard 1155: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1156: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1157: int popbased=0;
1158:
1159: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1160: int maxwav=0; /* Maxim number of waves */
1161: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1162: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1163: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1164: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1165: int mle=1, weightopt=0;
1.126 brouard 1166: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1167: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1168: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1169: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1170: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1171: int selected(int kvar); /* Is covariate kvar selected for printing results */
1172:
1.130 brouard 1173: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1174: double **matprod2(); /* test */
1.126 brouard 1175: double **oldm, **newm, **savm; /* Working pointers to matrices */
1176: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1177: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1178:
1.136 brouard 1179: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1180: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1181: FILE *ficlog, *ficrespow;
1.130 brouard 1182: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1183: double fretone; /* Only one call to likelihood */
1.130 brouard 1184: long ipmx=0; /* Number of contributions */
1.126 brouard 1185: double sw; /* Sum of weights */
1186: char filerespow[FILENAMELENGTH];
1187: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1188: FILE *ficresilk;
1189: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1190: FILE *ficresprobmorprev;
1191: FILE *fichtm, *fichtmcov; /* Html File */
1192: FILE *ficreseij;
1193: char filerese[FILENAMELENGTH];
1194: FILE *ficresstdeij;
1195: char fileresstde[FILENAMELENGTH];
1196: FILE *ficrescveij;
1197: char filerescve[FILENAMELENGTH];
1198: FILE *ficresvij;
1199: char fileresv[FILENAMELENGTH];
1.269 brouard 1200:
1.126 brouard 1201: char title[MAXLINE];
1.234 brouard 1202: char model[MAXLINE]; /**< The model line */
1.217 brouard 1203: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1204: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1205: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1206: char command[FILENAMELENGTH];
1207: int outcmd=0;
1208:
1.217 brouard 1209: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1210: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1211: char filelog[FILENAMELENGTH]; /* Log file */
1212: char filerest[FILENAMELENGTH];
1213: char fileregp[FILENAMELENGTH];
1214: char popfile[FILENAMELENGTH];
1215:
1216: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1217:
1.157 brouard 1218: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1219: /* struct timezone tzp; */
1220: /* extern int gettimeofday(); */
1221: struct tm tml, *gmtime(), *localtime();
1222:
1223: extern time_t time();
1224:
1225: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1226: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1227: struct tm tm;
1228:
1.126 brouard 1229: char strcurr[80], strfor[80];
1230:
1231: char *endptr;
1232: long lval;
1233: double dval;
1234:
1235: #define NR_END 1
1236: #define FREE_ARG char*
1237: #define FTOL 1.0e-10
1238:
1239: #define NRANSI
1.240 brouard 1240: #define ITMAX 200
1241: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1242:
1243: #define TOL 2.0e-4
1244:
1245: #define CGOLD 0.3819660
1246: #define ZEPS 1.0e-10
1247: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1248:
1249: #define GOLD 1.618034
1250: #define GLIMIT 100.0
1251: #define TINY 1.0e-20
1252:
1253: static double maxarg1,maxarg2;
1254: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1255: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1256:
1257: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1258: #define rint(a) floor(a+0.5)
1.166 brouard 1259: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1260: #define mytinydouble 1.0e-16
1.166 brouard 1261: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1262: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1263: /* static double dsqrarg; */
1264: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1265: static double sqrarg;
1266: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1267: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1268: int agegomp= AGEGOMP;
1269:
1270: int imx;
1271: int stepm=1;
1272: /* Stepm, step in month: minimum step interpolation*/
1273:
1274: int estepm;
1275: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1276:
1277: int m,nb;
1278: long *num;
1.197 brouard 1279: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1280: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1281: covariate for which somebody answered excluding
1282: undefined. Usually 2: 0 and 1. */
1283: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1284: covariate for which somebody answered including
1285: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1286: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1287: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1288: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1289: double *ageexmed,*agecens;
1290: double dateintmean=0;
1.296 brouard 1291: double anprojd, mprojd, jprojd; /* For eventual projections */
1292: double anprojf, mprojf, jprojf;
1.126 brouard 1293:
1.296 brouard 1294: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1295: double anbackf, mbackf, jbackf;
1296: double jintmean,mintmean,aintmean;
1.126 brouard 1297: double *weight;
1298: int **s; /* Status */
1.141 brouard 1299: double *agedc;
1.145 brouard 1300: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1301: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1302: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1303: double **coqvar; /* Fixed quantitative covariate nqv */
1304: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1305: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1306: double idx;
1307: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1308: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1309: /*k 1 2 3 4 5 6 7 8 9 */
1310: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1311: /* Tndvar[k] 1 2 3 4 5 */
1312: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1313: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1314: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1315: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1316: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1317: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1318: /* Tprod[i]=k 4 7 */
1319: /* Tage[i]=k 5 8 */
1320: /* */
1321: /* Type */
1322: /* V 1 2 3 4 5 */
1323: /* F F V V V */
1324: /* D Q D D Q */
1325: /* */
1326: int *TvarsD;
1327: int *TvarsDind;
1328: int *TvarsQ;
1329: int *TvarsQind;
1330:
1.235 brouard 1331: #define MAXRESULTLINES 10
1332: int nresult=0;
1.258 brouard 1333: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1334: int TKresult[MAXRESULTLINES];
1.237 brouard 1335: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1336: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1337: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1338: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1339: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1340: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1341:
1.234 brouard 1342: /* 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 1343: 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 */
1344: 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 */
1345: 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 */
1346: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1347: 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 */
1348: 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 1349: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1350: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1351: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1352: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1353: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1354: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1355: 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 */
1356: 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 */
1357:
1.230 brouard 1358: int *Tvarsel; /**< Selected covariates for output */
1359: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1360: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1361: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1362: 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 1363: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1364: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1365: int *Tage;
1.227 brouard 1366: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1367: 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 1368: 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*/
1369: 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 1370: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1371: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1372: int **Tvard;
1373: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1374: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1375: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1376: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1377: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1378: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1379: double *lsurv, *lpop, *tpop;
1380:
1.231 brouard 1381: #define FD 1; /* Fixed dummy covariate */
1382: #define FQ 2; /* Fixed quantitative covariate */
1383: #define FP 3; /* Fixed product covariate */
1384: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1385: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1386: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1387: #define VD 10; /* Varying dummy covariate */
1388: #define VQ 11; /* Varying quantitative covariate */
1389: #define VP 12; /* Varying product covariate */
1390: #define VPDD 13; /* Varying product dummy*dummy covariate */
1391: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1392: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1393: #define APFD 16; /* Age product * fixed dummy covariate */
1394: #define APFQ 17; /* Age product * fixed quantitative covariate */
1395: #define APVD 18; /* Age product * varying dummy covariate */
1396: #define APVQ 19; /* Age product * varying quantitative covariate */
1397:
1398: #define FTYPE 1; /* Fixed covariate */
1399: #define VTYPE 2; /* Varying covariate (loop in wave) */
1400: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1401:
1402: struct kmodel{
1403: int maintype; /* main type */
1404: int subtype; /* subtype */
1405: };
1406: struct kmodel modell[NCOVMAX];
1407:
1.143 brouard 1408: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1409: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1410:
1411: /**************** split *************************/
1412: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1413: {
1414: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1415: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1416: */
1417: char *ss; /* pointer */
1.186 brouard 1418: int l1=0, l2=0; /* length counters */
1.126 brouard 1419:
1420: l1 = strlen(path ); /* length of path */
1421: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1422: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1423: if ( ss == NULL ) { /* no directory, so determine current directory */
1424: strcpy( name, path ); /* we got the fullname name because no directory */
1425: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1426: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1427: /* get current working directory */
1428: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1429: #ifdef WIN32
1430: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1431: #else
1432: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1433: #endif
1.126 brouard 1434: return( GLOCK_ERROR_GETCWD );
1435: }
1436: /* got dirc from getcwd*/
1437: printf(" DIRC = %s \n",dirc);
1.205 brouard 1438: } else { /* strip directory from path */
1.126 brouard 1439: ss++; /* after this, the filename */
1440: l2 = strlen( ss ); /* length of filename */
1441: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1442: strcpy( name, ss ); /* save file name */
1443: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1444: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1445: printf(" DIRC2 = %s \n",dirc);
1446: }
1447: /* We add a separator at the end of dirc if not exists */
1448: l1 = strlen( dirc ); /* length of directory */
1449: if( dirc[l1-1] != DIRSEPARATOR ){
1450: dirc[l1] = DIRSEPARATOR;
1451: dirc[l1+1] = 0;
1452: printf(" DIRC3 = %s \n",dirc);
1453: }
1454: ss = strrchr( name, '.' ); /* find last / */
1455: if (ss >0){
1456: ss++;
1457: strcpy(ext,ss); /* save extension */
1458: l1= strlen( name);
1459: l2= strlen(ss)+1;
1460: strncpy( finame, name, l1-l2);
1461: finame[l1-l2]= 0;
1462: }
1463:
1464: return( 0 ); /* we're done */
1465: }
1466:
1467:
1468: /******************************************/
1469:
1470: void replace_back_to_slash(char *s, char*t)
1471: {
1472: int i;
1473: int lg=0;
1474: i=0;
1475: lg=strlen(t);
1476: for(i=0; i<= lg; i++) {
1477: (s[i] = t[i]);
1478: if (t[i]== '\\') s[i]='/';
1479: }
1480: }
1481:
1.132 brouard 1482: char *trimbb(char *out, char *in)
1.137 brouard 1483: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1484: char *s;
1485: s=out;
1486: while (*in != '\0'){
1.137 brouard 1487: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1488: in++;
1489: }
1490: *out++ = *in++;
1491: }
1492: *out='\0';
1493: return s;
1494: }
1495:
1.187 brouard 1496: /* char *substrchaine(char *out, char *in, char *chain) */
1497: /* { */
1498: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1499: /* char *s, *t; */
1500: /* t=in;s=out; */
1501: /* while ((*in != *chain) && (*in != '\0')){ */
1502: /* *out++ = *in++; */
1503: /* } */
1504:
1505: /* /\* *in matches *chain *\/ */
1506: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1507: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1508: /* } */
1509: /* in--; chain--; */
1510: /* while ( (*in != '\0')){ */
1511: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1512: /* *out++ = *in++; */
1513: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1514: /* } */
1515: /* *out='\0'; */
1516: /* out=s; */
1517: /* return out; */
1518: /* } */
1519: char *substrchaine(char *out, char *in, char *chain)
1520: {
1521: /* Substract chain 'chain' from 'in', return and output 'out' */
1522: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1523:
1524: char *strloc;
1525:
1526: strcpy (out, in);
1527: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1528: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1529: if(strloc != NULL){
1530: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1531: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1532: /* strcpy (strloc, strloc +strlen(chain));*/
1533: }
1534: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1535: return out;
1536: }
1537:
1538:
1.145 brouard 1539: char *cutl(char *blocc, char *alocc, char *in, char occ)
1540: {
1.187 brouard 1541: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1542: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1543: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1544: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1545: */
1.160 brouard 1546: char *s, *t;
1.145 brouard 1547: t=in;s=in;
1548: while ((*in != occ) && (*in != '\0')){
1549: *alocc++ = *in++;
1550: }
1551: if( *in == occ){
1552: *(alocc)='\0';
1553: s=++in;
1554: }
1555:
1556: if (s == t) {/* occ not found */
1557: *(alocc-(in-s))='\0';
1558: in=s;
1559: }
1560: while ( *in != '\0'){
1561: *blocc++ = *in++;
1562: }
1563:
1564: *blocc='\0';
1565: return t;
1566: }
1.137 brouard 1567: char *cutv(char *blocc, char *alocc, char *in, char occ)
1568: {
1.187 brouard 1569: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1570: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1571: gives blocc="abcdef2ghi" and alocc="j".
1572: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1573: */
1574: char *s, *t;
1575: t=in;s=in;
1576: while (*in != '\0'){
1577: while( *in == occ){
1578: *blocc++ = *in++;
1579: s=in;
1580: }
1581: *blocc++ = *in++;
1582: }
1583: if (s == t) /* occ not found */
1584: *(blocc-(in-s))='\0';
1585: else
1586: *(blocc-(in-s)-1)='\0';
1587: in=s;
1588: while ( *in != '\0'){
1589: *alocc++ = *in++;
1590: }
1591:
1592: *alocc='\0';
1593: return s;
1594: }
1595:
1.126 brouard 1596: int nbocc(char *s, char occ)
1597: {
1598: int i,j=0;
1599: int lg=20;
1600: i=0;
1601: lg=strlen(s);
1602: for(i=0; i<= lg; i++) {
1.234 brouard 1603: if (s[i] == occ ) j++;
1.126 brouard 1604: }
1605: return j;
1606: }
1607:
1.137 brouard 1608: /* void cutv(char *u,char *v, char*t, char occ) */
1609: /* { */
1610: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1611: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1612: /* gives u="abcdef2ghi" and v="j" *\/ */
1613: /* int i,lg,j,p=0; */
1614: /* i=0; */
1615: /* lg=strlen(t); */
1616: /* for(j=0; j<=lg-1; j++) { */
1617: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1618: /* } */
1.126 brouard 1619:
1.137 brouard 1620: /* for(j=0; j<p; j++) { */
1621: /* (u[j] = t[j]); */
1622: /* } */
1623: /* u[p]='\0'; */
1.126 brouard 1624:
1.137 brouard 1625: /* for(j=0; j<= lg; j++) { */
1626: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1627: /* } */
1628: /* } */
1.126 brouard 1629:
1.160 brouard 1630: #ifdef _WIN32
1631: char * strsep(char **pp, const char *delim)
1632: {
1633: char *p, *q;
1634:
1635: if ((p = *pp) == NULL)
1636: return 0;
1637: if ((q = strpbrk (p, delim)) != NULL)
1638: {
1639: *pp = q + 1;
1640: *q = '\0';
1641: }
1642: else
1643: *pp = 0;
1644: return p;
1645: }
1646: #endif
1647:
1.126 brouard 1648: /********************** nrerror ********************/
1649:
1650: void nrerror(char error_text[])
1651: {
1652: fprintf(stderr,"ERREUR ...\n");
1653: fprintf(stderr,"%s\n",error_text);
1654: exit(EXIT_FAILURE);
1655: }
1656: /*********************** vector *******************/
1657: double *vector(int nl, int nh)
1658: {
1659: double *v;
1660: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1661: if (!v) nrerror("allocation failure in vector");
1662: return v-nl+NR_END;
1663: }
1664:
1665: /************************ free vector ******************/
1666: void free_vector(double*v, int nl, int nh)
1667: {
1668: free((FREE_ARG)(v+nl-NR_END));
1669: }
1670:
1671: /************************ivector *******************************/
1672: int *ivector(long nl,long nh)
1673: {
1674: int *v;
1675: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1676: if (!v) nrerror("allocation failure in ivector");
1677: return v-nl+NR_END;
1678: }
1679:
1680: /******************free ivector **************************/
1681: void free_ivector(int *v, long nl, long nh)
1682: {
1683: free((FREE_ARG)(v+nl-NR_END));
1684: }
1685:
1686: /************************lvector *******************************/
1687: long *lvector(long nl,long nh)
1688: {
1689: long *v;
1690: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1691: if (!v) nrerror("allocation failure in ivector");
1692: return v-nl+NR_END;
1693: }
1694:
1695: /******************free lvector **************************/
1696: void free_lvector(long *v, long nl, long nh)
1697: {
1698: free((FREE_ARG)(v+nl-NR_END));
1699: }
1700:
1701: /******************* imatrix *******************************/
1702: int **imatrix(long nrl, long nrh, long ncl, long nch)
1703: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1704: {
1705: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1706: int **m;
1707:
1708: /* allocate pointers to rows */
1709: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1710: if (!m) nrerror("allocation failure 1 in matrix()");
1711: m += NR_END;
1712: m -= nrl;
1713:
1714:
1715: /* allocate rows and set pointers to them */
1716: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1717: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1718: m[nrl] += NR_END;
1719: m[nrl] -= ncl;
1720:
1721: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1722:
1723: /* return pointer to array of pointers to rows */
1724: return m;
1725: }
1726:
1727: /****************** free_imatrix *************************/
1728: void free_imatrix(m,nrl,nrh,ncl,nch)
1729: int **m;
1730: long nch,ncl,nrh,nrl;
1731: /* free an int matrix allocated by imatrix() */
1732: {
1733: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1734: free((FREE_ARG) (m+nrl-NR_END));
1735: }
1736:
1737: /******************* matrix *******************************/
1738: double **matrix(long nrl, long nrh, long ncl, long nch)
1739: {
1740: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1741: double **m;
1742:
1743: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1744: if (!m) nrerror("allocation failure 1 in matrix()");
1745: m += NR_END;
1746: m -= nrl;
1747:
1748: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1749: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1750: m[nrl] += NR_END;
1751: m[nrl] -= ncl;
1752:
1753: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1754: return m;
1.145 brouard 1755: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1756: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1757: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1758: */
1759: }
1760:
1761: /*************************free matrix ************************/
1762: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1763: {
1764: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1765: free((FREE_ARG)(m+nrl-NR_END));
1766: }
1767:
1768: /******************* ma3x *******************************/
1769: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1770: {
1771: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1772: double ***m;
1773:
1774: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1775: if (!m) nrerror("allocation failure 1 in matrix()");
1776: m += NR_END;
1777: m -= nrl;
1778:
1779: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1780: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1781: m[nrl] += NR_END;
1782: m[nrl] -= ncl;
1783:
1784: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1785:
1786: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1787: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1788: m[nrl][ncl] += NR_END;
1789: m[nrl][ncl] -= nll;
1790: for (j=ncl+1; j<=nch; j++)
1791: m[nrl][j]=m[nrl][j-1]+nlay;
1792:
1793: for (i=nrl+1; i<=nrh; i++) {
1794: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1795: for (j=ncl+1; j<=nch; j++)
1796: m[i][j]=m[i][j-1]+nlay;
1797: }
1798: return m;
1799: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1800: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1801: */
1802: }
1803:
1804: /*************************free ma3x ************************/
1805: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1806: {
1807: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1808: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1809: free((FREE_ARG)(m+nrl-NR_END));
1810: }
1811:
1812: /*************** function subdirf ***********/
1813: char *subdirf(char fileres[])
1814: {
1815: /* Caution optionfilefiname is hidden */
1816: strcpy(tmpout,optionfilefiname);
1817: strcat(tmpout,"/"); /* Add to the right */
1818: strcat(tmpout,fileres);
1819: return tmpout;
1820: }
1821:
1822: /*************** function subdirf2 ***********/
1823: char *subdirf2(char fileres[], char *preop)
1824: {
1825:
1826: /* Caution optionfilefiname is hidden */
1827: strcpy(tmpout,optionfilefiname);
1828: strcat(tmpout,"/");
1829: strcat(tmpout,preop);
1830: strcat(tmpout,fileres);
1831: return tmpout;
1832: }
1833:
1834: /*************** function subdirf3 ***********/
1835: char *subdirf3(char fileres[], char *preop, char *preop2)
1836: {
1837:
1838: /* Caution optionfilefiname is hidden */
1839: strcpy(tmpout,optionfilefiname);
1840: strcat(tmpout,"/");
1841: strcat(tmpout,preop);
1842: strcat(tmpout,preop2);
1843: strcat(tmpout,fileres);
1844: return tmpout;
1845: }
1.213 brouard 1846:
1847: /*************** function subdirfext ***********/
1848: char *subdirfext(char fileres[], char *preop, char *postop)
1849: {
1850:
1851: strcpy(tmpout,preop);
1852: strcat(tmpout,fileres);
1853: strcat(tmpout,postop);
1854: return tmpout;
1855: }
1.126 brouard 1856:
1.213 brouard 1857: /*************** function subdirfext3 ***********/
1858: char *subdirfext3(char fileres[], char *preop, char *postop)
1859: {
1860:
1861: /* Caution optionfilefiname is hidden */
1862: strcpy(tmpout,optionfilefiname);
1863: strcat(tmpout,"/");
1864: strcat(tmpout,preop);
1865: strcat(tmpout,fileres);
1866: strcat(tmpout,postop);
1867: return tmpout;
1868: }
1869:
1.162 brouard 1870: char *asc_diff_time(long time_sec, char ascdiff[])
1871: {
1872: long sec_left, days, hours, minutes;
1873: days = (time_sec) / (60*60*24);
1874: sec_left = (time_sec) % (60*60*24);
1875: hours = (sec_left) / (60*60) ;
1876: sec_left = (sec_left) %(60*60);
1877: minutes = (sec_left) /60;
1878: sec_left = (sec_left) % (60);
1879: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1880: return ascdiff;
1881: }
1882:
1.126 brouard 1883: /***************** f1dim *************************/
1884: extern int ncom;
1885: extern double *pcom,*xicom;
1886: extern double (*nrfunc)(double []);
1887:
1888: double f1dim(double x)
1889: {
1890: int j;
1891: double f;
1892: double *xt;
1893:
1894: xt=vector(1,ncom);
1895: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1896: f=(*nrfunc)(xt);
1897: free_vector(xt,1,ncom);
1898: return f;
1899: }
1900:
1901: /*****************brent *************************/
1902: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1903: {
1904: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1905: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1906: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1907: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1908: * returned function value.
1909: */
1.126 brouard 1910: int iter;
1911: double a,b,d,etemp;
1.159 brouard 1912: double fu=0,fv,fw,fx;
1.164 brouard 1913: double ftemp=0.;
1.126 brouard 1914: double p,q,r,tol1,tol2,u,v,w,x,xm;
1915: double e=0.0;
1916:
1917: a=(ax < cx ? ax : cx);
1918: b=(ax > cx ? ax : cx);
1919: x=w=v=bx;
1920: fw=fv=fx=(*f)(x);
1921: for (iter=1;iter<=ITMAX;iter++) {
1922: xm=0.5*(a+b);
1923: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1924: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1925: printf(".");fflush(stdout);
1926: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1927: #ifdef DEBUGBRENT
1.126 brouard 1928: 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);
1929: 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);
1930: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1931: #endif
1932: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1933: *xmin=x;
1934: return fx;
1935: }
1936: ftemp=fu;
1937: if (fabs(e) > tol1) {
1938: r=(x-w)*(fx-fv);
1939: q=(x-v)*(fx-fw);
1940: p=(x-v)*q-(x-w)*r;
1941: q=2.0*(q-r);
1942: if (q > 0.0) p = -p;
1943: q=fabs(q);
1944: etemp=e;
1945: e=d;
1946: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1947: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1948: else {
1.224 brouard 1949: d=p/q;
1950: u=x+d;
1951: if (u-a < tol2 || b-u < tol2)
1952: d=SIGN(tol1,xm-x);
1.126 brouard 1953: }
1954: } else {
1955: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1956: }
1957: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1958: fu=(*f)(u);
1959: if (fu <= fx) {
1960: if (u >= x) a=x; else b=x;
1961: SHFT(v,w,x,u)
1.183 brouard 1962: SHFT(fv,fw,fx,fu)
1963: } else {
1964: if (u < x) a=u; else b=u;
1965: if (fu <= fw || w == x) {
1.224 brouard 1966: v=w;
1967: w=u;
1968: fv=fw;
1969: fw=fu;
1.183 brouard 1970: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1971: v=u;
1972: fv=fu;
1.183 brouard 1973: }
1974: }
1.126 brouard 1975: }
1976: nrerror("Too many iterations in brent");
1977: *xmin=x;
1978: return fx;
1979: }
1980:
1981: /****************** mnbrak ***********************/
1982:
1983: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1984: double (*func)(double))
1.183 brouard 1985: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1986: the downhill direction (defined by the function as evaluated at the initial points) and returns
1987: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1988: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1989: */
1.126 brouard 1990: double ulim,u,r,q, dum;
1991: double fu;
1.187 brouard 1992:
1993: double scale=10.;
1994: int iterscale=0;
1995:
1996: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1997: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1998:
1999:
2000: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2001: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2002: /* *bx = *ax - (*ax - *bx)/scale; */
2003: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2004: /* } */
2005:
1.126 brouard 2006: if (*fb > *fa) {
2007: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2008: SHFT(dum,*fb,*fa,dum)
2009: }
1.126 brouard 2010: *cx=(*bx)+GOLD*(*bx-*ax);
2011: *fc=(*func)(*cx);
1.183 brouard 2012: #ifdef DEBUG
1.224 brouard 2013: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2014: 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 2015: #endif
1.224 brouard 2016: 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 2017: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2018: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2019: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2020: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2021: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2022: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2023: fu=(*func)(u);
1.163 brouard 2024: #ifdef DEBUG
2025: /* f(x)=A(x-u)**2+f(u) */
2026: double A, fparabu;
2027: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2028: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2029: 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);
2030: 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 2031: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2032: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2033: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2034: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2035: #endif
1.184 brouard 2036: #ifdef MNBRAKORIGINAL
1.183 brouard 2037: #else
1.191 brouard 2038: /* if (fu > *fc) { */
2039: /* #ifdef DEBUG */
2040: /* printf("mnbrak4 fu > fc \n"); */
2041: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2042: /* #endif */
2043: /* /\* 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 *\\/ *\/ */
2044: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2045: /* dum=u; /\* Shifting c and u *\/ */
2046: /* u = *cx; */
2047: /* *cx = dum; */
2048: /* dum = fu; */
2049: /* fu = *fc; */
2050: /* *fc =dum; */
2051: /* } else { /\* end *\/ */
2052: /* #ifdef DEBUG */
2053: /* printf("mnbrak3 fu < fc \n"); */
2054: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2055: /* #endif */
2056: /* dum=u; /\* Shifting c and u *\/ */
2057: /* u = *cx; */
2058: /* *cx = dum; */
2059: /* dum = fu; */
2060: /* fu = *fc; */
2061: /* *fc =dum; */
2062: /* } */
1.224 brouard 2063: #ifdef DEBUGMNBRAK
2064: double A, fparabu;
2065: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2066: fparabu= *fa - A*(*ax-u)*(*ax-u);
2067: 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);
2068: 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 2069: #endif
1.191 brouard 2070: dum=u; /* Shifting c and u */
2071: u = *cx;
2072: *cx = dum;
2073: dum = fu;
2074: fu = *fc;
2075: *fc =dum;
1.183 brouard 2076: #endif
1.162 brouard 2077: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2078: #ifdef DEBUG
1.224 brouard 2079: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2080: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2081: #endif
1.126 brouard 2082: fu=(*func)(u);
2083: if (fu < *fc) {
1.183 brouard 2084: #ifdef DEBUG
1.224 brouard 2085: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2086: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2087: #endif
2088: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2089: SHFT(*fb,*fc,fu,(*func)(u))
2090: #ifdef DEBUG
2091: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2092: #endif
2093: }
1.162 brouard 2094: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2095: #ifdef DEBUG
1.224 brouard 2096: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2097: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2098: #endif
1.126 brouard 2099: u=ulim;
2100: fu=(*func)(u);
1.183 brouard 2101: } else { /* u could be left to b (if r > q parabola has a maximum) */
2102: #ifdef DEBUG
1.224 brouard 2103: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2104: 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 2105: #endif
1.126 brouard 2106: u=(*cx)+GOLD*(*cx-*bx);
2107: fu=(*func)(u);
1.224 brouard 2108: #ifdef DEBUG
2109: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2110: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2111: #endif
1.183 brouard 2112: } /* end tests */
1.126 brouard 2113: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2114: SHFT(*fa,*fb,*fc,fu)
2115: #ifdef DEBUG
1.224 brouard 2116: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2117: 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 2118: #endif
2119: } /* 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 2120: }
2121:
2122: /*************** linmin ************************/
1.162 brouard 2123: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2124: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2125: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2126: the value of func at the returned location p . This is actually all accomplished by calling the
2127: routines mnbrak and brent .*/
1.126 brouard 2128: int ncom;
2129: double *pcom,*xicom;
2130: double (*nrfunc)(double []);
2131:
1.224 brouard 2132: #ifdef LINMINORIGINAL
1.126 brouard 2133: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2134: #else
2135: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2136: #endif
1.126 brouard 2137: {
2138: double brent(double ax, double bx, double cx,
2139: double (*f)(double), double tol, double *xmin);
2140: double f1dim(double x);
2141: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2142: double *fc, double (*func)(double));
2143: int j;
2144: double xx,xmin,bx,ax;
2145: double fx,fb,fa;
1.187 brouard 2146:
1.203 brouard 2147: #ifdef LINMINORIGINAL
2148: #else
2149: double scale=10., axs, xxs; /* Scale added for infinity */
2150: #endif
2151:
1.126 brouard 2152: ncom=n;
2153: pcom=vector(1,n);
2154: xicom=vector(1,n);
2155: nrfunc=func;
2156: for (j=1;j<=n;j++) {
2157: pcom[j]=p[j];
1.202 brouard 2158: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2159: }
1.187 brouard 2160:
1.203 brouard 2161: #ifdef LINMINORIGINAL
2162: xx=1.;
2163: #else
2164: axs=0.0;
2165: xxs=1.;
2166: do{
2167: xx= xxs;
2168: #endif
1.187 brouard 2169: ax=0.;
2170: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2171: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2172: /* 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)) */
2173: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2174: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2175: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2176: /* 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 2177: #ifdef LINMINORIGINAL
2178: #else
2179: if (fx != fx){
1.224 brouard 2180: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2181: printf("|");
2182: fprintf(ficlog,"|");
1.203 brouard 2183: #ifdef DEBUGLINMIN
1.224 brouard 2184: 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 2185: #endif
2186: }
1.224 brouard 2187: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2188: #endif
2189:
1.191 brouard 2190: #ifdef DEBUGLINMIN
2191: 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 2192: 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 2193: #endif
1.224 brouard 2194: #ifdef LINMINORIGINAL
2195: #else
2196: if(fb == fx){ /* Flat function in the direction */
2197: xmin=xx;
2198: *flat=1;
2199: }else{
2200: *flat=0;
2201: #endif
2202: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2203: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2204: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2205: /* fmin = f(p[j] + xmin * xi[j]) */
2206: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2207: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2208: #ifdef DEBUG
1.224 brouard 2209: 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);
2210: 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);
2211: #endif
2212: #ifdef LINMINORIGINAL
2213: #else
2214: }
1.126 brouard 2215: #endif
1.191 brouard 2216: #ifdef DEBUGLINMIN
2217: printf("linmin end ");
1.202 brouard 2218: fprintf(ficlog,"linmin end ");
1.191 brouard 2219: #endif
1.126 brouard 2220: for (j=1;j<=n;j++) {
1.203 brouard 2221: #ifdef LINMINORIGINAL
2222: xi[j] *= xmin;
2223: #else
2224: #ifdef DEBUGLINMIN
2225: if(xxs <1.0)
2226: printf(" before xi[%d]=%12.8f", j,xi[j]);
2227: #endif
2228: 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) */
2229: #ifdef DEBUGLINMIN
2230: if(xxs <1.0)
2231: 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 );
2232: #endif
2233: #endif
1.187 brouard 2234: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2235: }
1.191 brouard 2236: #ifdef DEBUGLINMIN
1.203 brouard 2237: printf("\n");
1.191 brouard 2238: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2239: 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 2240: for (j=1;j<=n;j++) {
1.202 brouard 2241: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2242: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2243: if(j % ncovmodel == 0){
1.191 brouard 2244: printf("\n");
1.202 brouard 2245: fprintf(ficlog,"\n");
2246: }
1.191 brouard 2247: }
1.203 brouard 2248: #else
1.191 brouard 2249: #endif
1.126 brouard 2250: free_vector(xicom,1,n);
2251: free_vector(pcom,1,n);
2252: }
2253:
2254:
2255: /*************** powell ************************/
1.162 brouard 2256: /*
2257: Minimization of a function func of n variables. Input consists of an initial starting point
2258: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2259: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2260: such that failure to decrease by more than this amount on one iteration signals doneness. On
2261: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2262: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2263: */
1.224 brouard 2264: #ifdef LINMINORIGINAL
2265: #else
2266: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2267: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2268: #endif
1.126 brouard 2269: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2270: double (*func)(double []))
2271: {
1.224 brouard 2272: #ifdef LINMINORIGINAL
2273: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2274: double (*func)(double []));
1.224 brouard 2275: #else
1.241 brouard 2276: void linmin(double p[], double xi[], int n, double *fret,
2277: double (*func)(double []),int *flat);
1.224 brouard 2278: #endif
1.239 brouard 2279: int i,ibig,j,jk,k;
1.126 brouard 2280: double del,t,*pt,*ptt,*xit;
1.181 brouard 2281: double directest;
1.126 brouard 2282: double fp,fptt;
2283: double *xits;
2284: int niterf, itmp;
1.224 brouard 2285: #ifdef LINMINORIGINAL
2286: #else
2287:
2288: flatdir=ivector(1,n);
2289: for (j=1;j<=n;j++) flatdir[j]=0;
2290: #endif
1.126 brouard 2291:
2292: pt=vector(1,n);
2293: ptt=vector(1,n);
2294: xit=vector(1,n);
2295: xits=vector(1,n);
2296: *fret=(*func)(p);
2297: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2298: rcurr_time = time(NULL);
1.126 brouard 2299: for (*iter=1;;++(*iter)) {
1.187 brouard 2300: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2301: ibig=0;
2302: del=0.0;
1.157 brouard 2303: rlast_time=rcurr_time;
2304: /* (void) gettimeofday(&curr_time,&tzp); */
2305: rcurr_time = time(NULL);
2306: curr_time = *localtime(&rcurr_time);
2307: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2308: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2309: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2310: for (i=1;i<=n;i++) {
1.126 brouard 2311: fprintf(ficrespow," %.12lf", p[i]);
2312: }
1.239 brouard 2313: fprintf(ficrespow,"\n");fflush(ficrespow);
2314: printf("\n#model= 1 + age ");
2315: fprintf(ficlog,"\n#model= 1 + age ");
2316: if(nagesqr==1){
1.241 brouard 2317: printf(" + age*age ");
2318: fprintf(ficlog," + age*age ");
1.239 brouard 2319: }
2320: for(j=1;j <=ncovmodel-2;j++){
2321: if(Typevar[j]==0) {
2322: printf(" + V%d ",Tvar[j]);
2323: fprintf(ficlog," + V%d ",Tvar[j]);
2324: }else if(Typevar[j]==1) {
2325: printf(" + V%d*age ",Tvar[j]);
2326: fprintf(ficlog," + V%d*age ",Tvar[j]);
2327: }else if(Typevar[j]==2) {
2328: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2329: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2330: }
2331: }
1.126 brouard 2332: printf("\n");
1.239 brouard 2333: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2334: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2335: fprintf(ficlog,"\n");
1.239 brouard 2336: for(i=1,jk=1; i <=nlstate; i++){
2337: for(k=1; k <=(nlstate+ndeath); k++){
2338: if (k != i) {
2339: printf("%d%d ",i,k);
2340: fprintf(ficlog,"%d%d ",i,k);
2341: for(j=1; j <=ncovmodel; j++){
2342: printf("%12.7f ",p[jk]);
2343: fprintf(ficlog,"%12.7f ",p[jk]);
2344: jk++;
2345: }
2346: printf("\n");
2347: fprintf(ficlog,"\n");
2348: }
2349: }
2350: }
1.241 brouard 2351: if(*iter <=3 && *iter >1){
1.157 brouard 2352: tml = *localtime(&rcurr_time);
2353: strcpy(strcurr,asctime(&tml));
2354: rforecast_time=rcurr_time;
1.126 brouard 2355: itmp = strlen(strcurr);
2356: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2357: strcurr[itmp-1]='\0';
1.162 brouard 2358: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2359: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2360: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2361: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2362: forecast_time = *localtime(&rforecast_time);
2363: strcpy(strfor,asctime(&forecast_time));
2364: itmp = strlen(strfor);
2365: if(strfor[itmp-1]=='\n')
2366: strfor[itmp-1]='\0';
2367: 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);
2368: 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 2369: }
2370: }
1.187 brouard 2371: for (i=1;i<=n;i++) { /* For each direction i */
2372: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2373: fptt=(*fret);
2374: #ifdef DEBUG
1.203 brouard 2375: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2376: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2377: #endif
1.203 brouard 2378: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2379: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2380: #ifdef LINMINORIGINAL
1.188 brouard 2381: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2382: #else
2383: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2384: flatdir[i]=flat; /* Function is vanishing in that direction i */
2385: #endif
2386: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2387: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2388: /* because that direction will be replaced unless the gain del is small */
2389: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2390: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2391: /* with the new direction. */
2392: del=fabs(fptt-(*fret));
2393: ibig=i;
1.126 brouard 2394: }
2395: #ifdef DEBUG
2396: printf("%d %.12e",i,(*fret));
2397: fprintf(ficlog,"%d %.12e",i,(*fret));
2398: for (j=1;j<=n;j++) {
1.224 brouard 2399: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2400: printf(" x(%d)=%.12e",j,xit[j]);
2401: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2402: }
2403: for(j=1;j<=n;j++) {
1.225 brouard 2404: printf(" p(%d)=%.12e",j,p[j]);
2405: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2406: }
2407: printf("\n");
2408: fprintf(ficlog,"\n");
2409: #endif
1.187 brouard 2410: } /* end loop on each direction i */
2411: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2412: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2413: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2414: for(j=1;j<=n;j++) {
1.225 brouard 2415: if(flatdir[j] >0){
2416: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2417: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2418: }
2419: /* printf("\n"); */
2420: /* fprintf(ficlog,"\n"); */
2421: }
1.243 brouard 2422: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2423: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2424: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2425: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2426: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2427: /* decreased of more than 3.84 */
2428: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2429: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2430: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2431:
1.188 brouard 2432: /* Starting the program with initial values given by a former maximization will simply change */
2433: /* the scales of the directions and the directions, because the are reset to canonical directions */
2434: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2435: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2436: #ifdef DEBUG
2437: int k[2],l;
2438: k[0]=1;
2439: k[1]=-1;
2440: printf("Max: %.12e",(*func)(p));
2441: fprintf(ficlog,"Max: %.12e",(*func)(p));
2442: for (j=1;j<=n;j++) {
2443: printf(" %.12e",p[j]);
2444: fprintf(ficlog," %.12e",p[j]);
2445: }
2446: printf("\n");
2447: fprintf(ficlog,"\n");
2448: for(l=0;l<=1;l++) {
2449: for (j=1;j<=n;j++) {
2450: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2451: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2452: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2453: }
2454: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2455: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2456: }
2457: #endif
2458:
1.224 brouard 2459: #ifdef LINMINORIGINAL
2460: #else
2461: free_ivector(flatdir,1,n);
2462: #endif
1.126 brouard 2463: free_vector(xit,1,n);
2464: free_vector(xits,1,n);
2465: free_vector(ptt,1,n);
2466: free_vector(pt,1,n);
2467: return;
1.192 brouard 2468: } /* enough precision */
1.240 brouard 2469: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2470: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2471: ptt[j]=2.0*p[j]-pt[j];
2472: xit[j]=p[j]-pt[j];
2473: pt[j]=p[j];
2474: }
1.181 brouard 2475: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2476: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2477: if (*iter <=4) {
1.225 brouard 2478: #else
2479: #endif
1.224 brouard 2480: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2481: #else
1.161 brouard 2482: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2483: #endif
1.162 brouard 2484: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2485: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2486: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2487: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2488: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2489: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2490: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2491: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2492: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2493: /* Even if f3 <f1, directest can be negative and t >0 */
2494: /* mu² and del² are equal when f3=f1 */
2495: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2496: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2497: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2498: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2499: #ifdef NRCORIGINAL
2500: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2501: #else
2502: 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 2503: t= t- del*SQR(fp-fptt);
1.183 brouard 2504: #endif
1.202 brouard 2505: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2506: #ifdef DEBUG
1.181 brouard 2507: 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);
2508: 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 2509: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2510: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2511: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2512: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2513: 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);
2514: 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);
2515: #endif
1.183 brouard 2516: #ifdef POWELLORIGINAL
2517: if (t < 0.0) { /* Then we use it for new direction */
2518: #else
1.182 brouard 2519: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2520: 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 2521: 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 2522: 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 2523: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2524: }
1.181 brouard 2525: if (directest < 0.0) { /* Then we use it for new direction */
2526: #endif
1.191 brouard 2527: #ifdef DEBUGLINMIN
1.234 brouard 2528: printf("Before linmin in direction P%d-P0\n",n);
2529: for (j=1;j<=n;j++) {
2530: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2531: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2532: if(j % ncovmodel == 0){
2533: printf("\n");
2534: fprintf(ficlog,"\n");
2535: }
2536: }
1.224 brouard 2537: #endif
2538: #ifdef LINMINORIGINAL
1.234 brouard 2539: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2540: #else
1.234 brouard 2541: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2542: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2543: #endif
1.234 brouard 2544:
1.191 brouard 2545: #ifdef DEBUGLINMIN
1.234 brouard 2546: for (j=1;j<=n;j++) {
2547: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2548: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2549: if(j % ncovmodel == 0){
2550: printf("\n");
2551: fprintf(ficlog,"\n");
2552: }
2553: }
1.224 brouard 2554: #endif
1.234 brouard 2555: for (j=1;j<=n;j++) {
2556: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2557: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2558: }
1.224 brouard 2559: #ifdef LINMINORIGINAL
2560: #else
1.234 brouard 2561: for (j=1, flatd=0;j<=n;j++) {
2562: if(flatdir[j]>0)
2563: flatd++;
2564: }
2565: if(flatd >0){
1.255 brouard 2566: printf("%d flat directions: ",flatd);
2567: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2568: for (j=1;j<=n;j++) {
2569: if(flatdir[j]>0){
2570: printf("%d ",j);
2571: fprintf(ficlog,"%d ",j);
2572: }
2573: }
2574: printf("\n");
2575: fprintf(ficlog,"\n");
2576: }
1.191 brouard 2577: #endif
1.234 brouard 2578: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2579: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2580:
1.126 brouard 2581: #ifdef DEBUG
1.234 brouard 2582: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2583: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2584: for(j=1;j<=n;j++){
2585: printf(" %lf",xit[j]);
2586: fprintf(ficlog," %lf",xit[j]);
2587: }
2588: printf("\n");
2589: fprintf(ficlog,"\n");
1.126 brouard 2590: #endif
1.192 brouard 2591: } /* end of t or directest negative */
1.224 brouard 2592: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2593: #else
1.234 brouard 2594: } /* end if (fptt < fp) */
1.192 brouard 2595: #endif
1.225 brouard 2596: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2597: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2598: #else
1.224 brouard 2599: #endif
1.234 brouard 2600: } /* loop iteration */
1.126 brouard 2601: }
1.234 brouard 2602:
1.126 brouard 2603: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2604:
1.235 brouard 2605: 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 2606: {
1.279 brouard 2607: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2608: * (and selected quantitative values in nres)
2609: * by left multiplying the unit
2610: * matrix by transitions matrix until convergence is reached with precision ftolpl
2611: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2612: * Wx is row vector: population in state 1, population in state 2, population dead
2613: * or prevalence in state 1, prevalence in state 2, 0
2614: * newm is the matrix after multiplications, its rows are identical at a factor.
2615: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2616: * Output is prlim.
2617: * Initial matrix pimij
2618: */
1.206 brouard 2619: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2620: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2621: /* 0, 0 , 1} */
2622: /*
2623: * and after some iteration: */
2624: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2625: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2626: /* 0, 0 , 1} */
2627: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2628: /* {0.51571254859325999, 0.4842874514067399, */
2629: /* 0.51326036147820708, 0.48673963852179264} */
2630: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2631:
1.126 brouard 2632: int i, ii,j,k;
1.209 brouard 2633: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2634: /* double **matprod2(); */ /* test */
1.218 brouard 2635: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2636: double **newm;
1.209 brouard 2637: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2638: int ncvloop=0;
1.288 brouard 2639: int first=0;
1.169 brouard 2640:
1.209 brouard 2641: min=vector(1,nlstate);
2642: max=vector(1,nlstate);
2643: meandiff=vector(1,nlstate);
2644:
1.218 brouard 2645: /* Starting with matrix unity */
1.126 brouard 2646: for (ii=1;ii<=nlstate+ndeath;ii++)
2647: for (j=1;j<=nlstate+ndeath;j++){
2648: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2649: }
1.169 brouard 2650:
2651: cov[1]=1.;
2652:
2653: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2654: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2655: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2656: ncvloop++;
1.126 brouard 2657: newm=savm;
2658: /* Covariates have to be included here again */
1.138 brouard 2659: cov[2]=agefin;
1.187 brouard 2660: if(nagesqr==1)
2661: cov[3]= agefin*agefin;;
1.234 brouard 2662: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2663: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2664: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2665: /* 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 2666: }
2667: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2668: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2669: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2670: /* 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 2671: }
1.237 brouard 2672: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2673: if(Dummy[Tvar[Tage[k]]]){
2674: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2675: } else{
1.235 brouard 2676: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2677: }
1.235 brouard 2678: /* 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 2679: }
1.237 brouard 2680: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2681: /* 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 2682: if(Dummy[Tvard[k][1]==0]){
2683: if(Dummy[Tvard[k][2]==0]){
2684: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2685: }else{
2686: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2687: }
2688: }else{
2689: if(Dummy[Tvard[k][2]==0]){
2690: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2691: }else{
2692: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2693: }
2694: }
1.234 brouard 2695: }
1.138 brouard 2696: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2697: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2698: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2699: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2700: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2701: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2702: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2703:
1.126 brouard 2704: savm=oldm;
2705: oldm=newm;
1.209 brouard 2706:
2707: for(j=1; j<=nlstate; j++){
2708: max[j]=0.;
2709: min[j]=1.;
2710: }
2711: for(i=1;i<=nlstate;i++){
2712: sumnew=0;
2713: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2714: for(j=1; j<=nlstate; j++){
2715: prlim[i][j]= newm[i][j]/(1-sumnew);
2716: max[j]=FMAX(max[j],prlim[i][j]);
2717: min[j]=FMIN(min[j],prlim[i][j]);
2718: }
2719: }
2720:
1.126 brouard 2721: maxmax=0.;
1.209 brouard 2722: for(j=1; j<=nlstate; j++){
2723: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2724: maxmax=FMAX(maxmax,meandiff[j]);
2725: /* 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 2726: } /* j loop */
1.203 brouard 2727: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2728: /* 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 2729: if(maxmax < ftolpl){
1.209 brouard 2730: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2731: free_vector(min,1,nlstate);
2732: free_vector(max,1,nlstate);
2733: free_vector(meandiff,1,nlstate);
1.126 brouard 2734: return prlim;
2735: }
1.288 brouard 2736: } /* agefin loop */
1.208 brouard 2737: /* After some age loop it doesn't converge */
1.288 brouard 2738: if(!first){
2739: first=1;
2740: 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);
2741: }
2742: 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);
2743:
1.209 brouard 2744: /* 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); */
2745: free_vector(min,1,nlstate);
2746: free_vector(max,1,nlstate);
2747: free_vector(meandiff,1,nlstate);
1.208 brouard 2748:
1.169 brouard 2749: return prlim; /* should not reach here */
1.126 brouard 2750: }
2751:
1.217 brouard 2752:
2753: /**** Back Prevalence limit (stable or period prevalence) ****************/
2754:
1.218 brouard 2755: /* 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) */
2756: /* 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 2757: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2758: {
1.264 brouard 2759: /* 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 2760: matrix by transitions matrix until convergence is reached with precision ftolpl */
2761: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2762: /* Wx is row vector: population in state 1, population in state 2, population dead */
2763: /* or prevalence in state 1, prevalence in state 2, 0 */
2764: /* newm is the matrix after multiplications, its rows are identical at a factor */
2765: /* Initial matrix pimij */
2766: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2767: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2768: /* 0, 0 , 1} */
2769: /*
2770: * and after some iteration: */
2771: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2772: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2773: /* 0, 0 , 1} */
2774: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2775: /* {0.51571254859325999, 0.4842874514067399, */
2776: /* 0.51326036147820708, 0.48673963852179264} */
2777: /* If we start from prlim again, prlim tends to a constant matrix */
2778:
2779: int i, ii,j,k;
1.247 brouard 2780: int first=0;
1.217 brouard 2781: double *min, *max, *meandiff, maxmax,sumnew=0.;
2782: /* double **matprod2(); */ /* test */
2783: double **out, cov[NCOVMAX+1], **bmij();
2784: double **newm;
1.218 brouard 2785: double **dnewm, **doldm, **dsavm; /* for use */
2786: double **oldm, **savm; /* for use */
2787:
1.217 brouard 2788: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2789: int ncvloop=0;
2790:
2791: min=vector(1,nlstate);
2792: max=vector(1,nlstate);
2793: meandiff=vector(1,nlstate);
2794:
1.266 brouard 2795: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2796: oldm=oldms; savm=savms;
2797:
2798: /* Starting with matrix unity */
2799: for (ii=1;ii<=nlstate+ndeath;ii++)
2800: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2801: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2802: }
2803:
2804: cov[1]=1.;
2805:
2806: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2807: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2808: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2809: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2810: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2811: ncvloop++;
1.218 brouard 2812: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2813: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2814: /* Covariates have to be included here again */
2815: cov[2]=agefin;
2816: if(nagesqr==1)
2817: cov[3]= agefin*agefin;;
1.242 brouard 2818: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2819: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2820: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2821: /* 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 2822: }
2823: /* for (k=1; k<=cptcovn;k++) { */
2824: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2825: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2826: /* /\* 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])]); *\/ */
2827: /* } */
2828: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2829: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2830: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2831: /* 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]); */
2832: }
2833: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2834: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2835: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2836: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2837: for (k=1; k<=cptcovage;k++){ /* For product with age */
2838: if(Dummy[Tvar[Tage[k]]]){
2839: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2840: } else{
2841: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2842: }
2843: /* 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]); */
2844: }
2845: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2846: /* 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]); */
2847: if(Dummy[Tvard[k][1]==0]){
2848: if(Dummy[Tvard[k][2]==0]){
2849: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2850: }else{
2851: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2852: }
2853: }else{
2854: if(Dummy[Tvard[k][2]==0]){
2855: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2856: }else{
2857: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2858: }
2859: }
1.217 brouard 2860: }
2861:
2862: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2863: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2864: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2865: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2866: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2867: /* ij should be linked to the correct index of cov */
2868: /* age and covariate values ij are in 'cov', but we need to pass
2869: * ij for the observed prevalence at age and status and covariate
2870: * number: prevacurrent[(int)agefin][ii][ij]
2871: */
2872: /* 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 *\/ */
2873: /* 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 *\/ */
2874: 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 2875: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2876: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2877: /* for(i=1; i<=nlstate+ndeath; i++) { */
2878: /* printf("%d newm= ",i); */
2879: /* for(j=1;j<=nlstate+ndeath;j++) { */
2880: /* printf("%f ",newm[i][j]); */
2881: /* } */
2882: /* printf("oldm * "); */
2883: /* for(j=1;j<=nlstate+ndeath;j++) { */
2884: /* printf("%f ",oldm[i][j]); */
2885: /* } */
1.268 brouard 2886: /* printf(" bmmij "); */
1.266 brouard 2887: /* for(j=1;j<=nlstate+ndeath;j++) { */
2888: /* printf("%f ",pmmij[i][j]); */
2889: /* } */
2890: /* printf("\n"); */
2891: /* } */
2892: /* } */
1.217 brouard 2893: savm=oldm;
2894: oldm=newm;
1.266 brouard 2895:
1.217 brouard 2896: for(j=1; j<=nlstate; j++){
2897: max[j]=0.;
2898: min[j]=1.;
2899: }
2900: for(j=1; j<=nlstate; j++){
2901: for(i=1;i<=nlstate;i++){
1.234 brouard 2902: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2903: bprlim[i][j]= newm[i][j];
2904: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2905: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2906: }
2907: }
1.218 brouard 2908:
1.217 brouard 2909: maxmax=0.;
2910: for(i=1; i<=nlstate; i++){
2911: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2912: maxmax=FMAX(maxmax,meandiff[i]);
2913: /* 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 2914: } /* i loop */
1.217 brouard 2915: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2916: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2917: if(maxmax < ftolpl){
1.220 brouard 2918: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2919: free_vector(min,1,nlstate);
2920: free_vector(max,1,nlstate);
2921: free_vector(meandiff,1,nlstate);
2922: return bprlim;
2923: }
1.288 brouard 2924: } /* agefin loop */
1.217 brouard 2925: /* After some age loop it doesn't converge */
1.288 brouard 2926: if(!first){
1.247 brouard 2927: first=1;
2928: 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\
2929: 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);
2930: }
2931: 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 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: /* 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); */
2934: free_vector(min,1,nlstate);
2935: free_vector(max,1,nlstate);
2936: free_vector(meandiff,1,nlstate);
2937:
2938: return bprlim; /* should not reach here */
2939: }
2940:
1.126 brouard 2941: /*************** transition probabilities ***************/
2942:
2943: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2944: {
1.138 brouard 2945: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2946: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2947: model to the ncovmodel covariates (including constant and age).
2948: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2949: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2950: ncth covariate in the global vector x is given by the formula:
2951: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2952: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2953: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2954: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2955: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2956: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2957: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2958: */
2959: double s1, lnpijopii;
1.126 brouard 2960: /*double t34;*/
1.164 brouard 2961: int i,j, nc, ii, jj;
1.126 brouard 2962:
1.223 brouard 2963: for(i=1; i<= nlstate; i++){
2964: for(j=1; j<i;j++){
2965: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2966: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2967: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2968: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2969: }
2970: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2971: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2972: }
2973: for(j=i+1; j<=nlstate+ndeath;j++){
2974: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2975: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2976: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2977: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2978: }
2979: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2980: }
2981: }
1.218 brouard 2982:
1.223 brouard 2983: for(i=1; i<= nlstate; i++){
2984: s1=0;
2985: for(j=1; j<i; j++){
2986: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2987: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2988: }
2989: for(j=i+1; j<=nlstate+ndeath; j++){
2990: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2991: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2992: }
2993: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2994: ps[i][i]=1./(s1+1.);
2995: /* Computing other pijs */
2996: for(j=1; j<i; j++)
2997: ps[i][j]= exp(ps[i][j])*ps[i][i];
2998: for(j=i+1; j<=nlstate+ndeath; j++)
2999: ps[i][j]= exp(ps[i][j])*ps[i][i];
3000: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3001: } /* end i */
1.218 brouard 3002:
1.223 brouard 3003: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3004: for(jj=1; jj<= nlstate+ndeath; jj++){
3005: ps[ii][jj]=0;
3006: ps[ii][ii]=1;
3007: }
3008: }
1.294 brouard 3009:
3010:
1.223 brouard 3011: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3012: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3013: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3014: /* } */
3015: /* printf("\n "); */
3016: /* } */
3017: /* printf("\n ");printf("%lf ",cov[2]);*/
3018: /*
3019: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3020: goto end;*/
1.266 brouard 3021: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3022: }
3023:
1.218 brouard 3024: /*************** backward transition probabilities ***************/
3025:
3026: /* 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 ) */
3027: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3028: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3029: {
1.266 brouard 3030: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
3031: * 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 3032: */
1.218 brouard 3033: int i, ii, j,k;
1.222 brouard 3034:
3035: double **out, **pmij();
3036: double sumnew=0.;
1.218 brouard 3037: double agefin;
1.292 brouard 3038: 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 3039: double **dnewm, **dsavm, **doldm;
3040: double **bbmij;
3041:
1.218 brouard 3042: doldm=ddoldms; /* global pointers */
1.222 brouard 3043: dnewm=ddnewms;
3044: dsavm=ddsavms;
3045:
3046: agefin=cov[2];
1.268 brouard 3047: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3048: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3049: the observed prevalence (with this covariate ij) at beginning of transition */
3050: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3051:
3052: /* P_x */
1.266 brouard 3053: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3054: /* outputs pmmij which is a stochastic matrix in row */
3055:
3056: /* Diag(w_x) */
1.292 brouard 3057: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3058: sumnew=0.;
1.269 brouard 3059: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3060: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3061: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3062: sumnew+=prevacurrent[(int)agefin][ii][ij];
3063: }
3064: if(sumnew >0.01){ /* At least some value in the prevalence */
3065: for (ii=1;ii<=nlstate+ndeath;ii++){
3066: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3067: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3068: }
3069: }else{
3070: for (ii=1;ii<=nlstate+ndeath;ii++){
3071: for (j=1;j<=nlstate+ndeath;j++)
3072: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3073: }
3074: /* if(sumnew <0.9){ */
3075: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3076: /* } */
3077: }
3078: k3=0.0; /* We put the last diagonal to 0 */
3079: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3080: doldm[ii][ii]= k3;
3081: }
3082: /* End doldm, At the end doldm is diag[(w_i)] */
3083:
1.292 brouard 3084: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3085: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3086:
1.292 brouard 3087: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3088: /* 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 3089: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3090: sumnew=0.;
1.222 brouard 3091: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3092: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3093: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3094: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3095: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3096: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3097: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3098: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3099: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3100: /* }else */
1.268 brouard 3101: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3102: } /*End ii */
3103: } /* 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 */
3104:
1.292 brouard 3105: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3106: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3107: /* end bmij */
1.266 brouard 3108: return ps; /*pointer is unchanged */
1.218 brouard 3109: }
1.217 brouard 3110: /*************** transition probabilities ***************/
3111:
1.218 brouard 3112: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3113: {
3114: /* According to parameters values stored in x and the covariate's values stored in cov,
3115: computes the probability to be observed in state j being in state i by appying the
3116: model to the ncovmodel covariates (including constant and age).
3117: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3118: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3119: ncth covariate in the global vector x is given by the formula:
3120: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3121: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3122: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3123: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3124: Outputs ps[i][j] the probability to be observed in j being in j according to
3125: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3126: */
3127: double s1, lnpijopii;
3128: /*double t34;*/
3129: int i,j, nc, ii, jj;
3130:
1.234 brouard 3131: for(i=1; i<= nlstate; i++){
3132: for(j=1; j<i;j++){
3133: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3134: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3135: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3136: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3137: }
3138: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3139: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3140: }
3141: for(j=i+1; j<=nlstate+ndeath;j++){
3142: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3143: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3144: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3145: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3146: }
3147: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3148: }
3149: }
3150:
3151: for(i=1; i<= nlstate; i++){
3152: s1=0;
3153: for(j=1; j<i; j++){
3154: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3155: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3156: }
3157: for(j=i+1; j<=nlstate+ndeath; j++){
3158: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3159: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3160: }
3161: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3162: ps[i][i]=1./(s1+1.);
3163: /* Computing other pijs */
3164: for(j=1; j<i; j++)
3165: ps[i][j]= exp(ps[i][j])*ps[i][i];
3166: for(j=i+1; j<=nlstate+ndeath; j++)
3167: ps[i][j]= exp(ps[i][j])*ps[i][i];
3168: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3169: } /* end i */
3170:
3171: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3172: for(jj=1; jj<= nlstate+ndeath; jj++){
3173: ps[ii][jj]=0;
3174: ps[ii][ii]=1;
3175: }
3176: }
1.296 brouard 3177: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3178: for(jj=1; jj<= nlstate+ndeath; jj++){
3179: s1=0.;
3180: for(ii=1; ii<= nlstate+ndeath; ii++){
3181: s1+=ps[ii][jj];
3182: }
3183: for(ii=1; ii<= nlstate; ii++){
3184: ps[ii][jj]=ps[ii][jj]/s1;
3185: }
3186: }
3187: /* Transposition */
3188: for(jj=1; jj<= nlstate+ndeath; jj++){
3189: for(ii=jj; ii<= nlstate+ndeath; ii++){
3190: s1=ps[ii][jj];
3191: ps[ii][jj]=ps[jj][ii];
3192: ps[jj][ii]=s1;
3193: }
3194: }
3195: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3196: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3197: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3198: /* } */
3199: /* printf("\n "); */
3200: /* } */
3201: /* printf("\n ");printf("%lf ",cov[2]);*/
3202: /*
3203: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3204: goto end;*/
3205: return ps;
1.217 brouard 3206: }
3207:
3208:
1.126 brouard 3209: /**************** Product of 2 matrices ******************/
3210:
1.145 brouard 3211: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3212: {
3213: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3214: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3215: /* in, b, out are matrice of pointers which should have been initialized
3216: before: only the contents of out is modified. The function returns
3217: a pointer to pointers identical to out */
1.145 brouard 3218: int i, j, k;
1.126 brouard 3219: for(i=nrl; i<= nrh; i++)
1.145 brouard 3220: for(k=ncolol; k<=ncoloh; k++){
3221: out[i][k]=0.;
3222: for(j=ncl; j<=nch; j++)
3223: out[i][k] +=in[i][j]*b[j][k];
3224: }
1.126 brouard 3225: return out;
3226: }
3227:
3228:
3229: /************* Higher Matrix Product ***************/
3230:
1.235 brouard 3231: 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 3232: {
1.218 brouard 3233: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3234: 'nhstepm*hstepm*stepm' months (i.e. until
3235: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3236: nhstepm*hstepm matrices.
3237: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3238: (typically every 2 years instead of every month which is too big
3239: for the memory).
3240: Model is determined by parameters x and covariates have to be
3241: included manually here.
3242:
3243: */
3244:
3245: int i, j, d, h, k;
1.131 brouard 3246: double **out, cov[NCOVMAX+1];
1.126 brouard 3247: double **newm;
1.187 brouard 3248: double agexact;
1.214 brouard 3249: double agebegin, ageend;
1.126 brouard 3250:
3251: /* Hstepm could be zero and should return the unit matrix */
3252: for (i=1;i<=nlstate+ndeath;i++)
3253: for (j=1;j<=nlstate+ndeath;j++){
3254: oldm[i][j]=(i==j ? 1.0 : 0.0);
3255: po[i][j][0]=(i==j ? 1.0 : 0.0);
3256: }
3257: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3258: for(h=1; h <=nhstepm; h++){
3259: for(d=1; d <=hstepm; d++){
3260: newm=savm;
3261: /* Covariates have to be included here again */
3262: cov[1]=1.;
1.214 brouard 3263: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3264: cov[2]=agexact;
3265: if(nagesqr==1)
1.227 brouard 3266: cov[3]= agexact*agexact;
1.235 brouard 3267: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3268: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3269: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3270: /* 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)); */
3271: }
3272: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3273: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3274: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3275: /* 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]); */
3276: }
3277: for (k=1; k<=cptcovage;k++){
3278: if(Dummy[Tvar[Tage[k]]]){
3279: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3280: } else{
3281: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3282: }
3283: /* 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]); */
3284: }
3285: for (k=1; k<=cptcovprod;k++){ /* */
3286: /* 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]); */
3287: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3288: }
3289: /* for (k=1; k<=cptcovn;k++) */
3290: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3291: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3292: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3293: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3294: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3295:
3296:
1.126 brouard 3297: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3298: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3299: /* right multiplication of oldm by the current matrix */
1.126 brouard 3300: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3301: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3302: /* if((int)age == 70){ */
3303: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3304: /* for(i=1; i<=nlstate+ndeath; i++) { */
3305: /* printf("%d pmmij ",i); */
3306: /* for(j=1;j<=nlstate+ndeath;j++) { */
3307: /* printf("%f ",pmmij[i][j]); */
3308: /* } */
3309: /* printf(" oldm "); */
3310: /* for(j=1;j<=nlstate+ndeath;j++) { */
3311: /* printf("%f ",oldm[i][j]); */
3312: /* } */
3313: /* printf("\n"); */
3314: /* } */
3315: /* } */
1.126 brouard 3316: savm=oldm;
3317: oldm=newm;
3318: }
3319: for(i=1; i<=nlstate+ndeath; i++)
3320: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3321: po[i][j][h]=newm[i][j];
3322: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3323: }
1.128 brouard 3324: /*printf("h=%d ",h);*/
1.126 brouard 3325: } /* end h */
1.267 brouard 3326: /* printf("\n H=%d \n",h); */
1.126 brouard 3327: return po;
3328: }
3329:
1.217 brouard 3330: /************* Higher Back Matrix Product ***************/
1.218 brouard 3331: /* 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 3332: 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 3333: {
1.266 brouard 3334: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3335: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3336: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3337: nhstepm*hstepm matrices.
3338: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3339: (typically every 2 years instead of every month which is too big
1.217 brouard 3340: for the memory).
1.218 brouard 3341: Model is determined by parameters x and covariates have to be
1.266 brouard 3342: included manually here. Then we use a call to bmij(x and cov)
3343: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3344: */
1.217 brouard 3345:
3346: int i, j, d, h, k;
1.266 brouard 3347: double **out, cov[NCOVMAX+1], **bmij();
3348: double **newm, ***newmm;
1.217 brouard 3349: double agexact;
3350: double agebegin, ageend;
1.222 brouard 3351: double **oldm, **savm;
1.217 brouard 3352:
1.266 brouard 3353: newmm=po; /* To be saved */
3354: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3355: /* Hstepm could be zero and should return the unit matrix */
3356: for (i=1;i<=nlstate+ndeath;i++)
3357: for (j=1;j<=nlstate+ndeath;j++){
3358: oldm[i][j]=(i==j ? 1.0 : 0.0);
3359: po[i][j][0]=(i==j ? 1.0 : 0.0);
3360: }
3361: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3362: for(h=1; h <=nhstepm; h++){
3363: for(d=1; d <=hstepm; d++){
3364: newm=savm;
3365: /* Covariates have to be included here again */
3366: cov[1]=1.;
1.271 brouard 3367: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3368: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3369: cov[2]=agexact;
3370: if(nagesqr==1)
1.222 brouard 3371: cov[3]= agexact*agexact;
1.266 brouard 3372: for (k=1; k<=cptcovn;k++){
3373: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3374: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3375: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3376: /* 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)); */
3377: }
1.267 brouard 3378: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3379: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3380: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3381: /* 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]); */
3382: }
3383: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3384: if(Dummy[Tvar[Tage[k]]]){
3385: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3386: } else{
3387: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3388: }
3389: /* 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]); */
3390: }
3391: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3392: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3393: }
1.217 brouard 3394: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3395: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3396:
1.218 brouard 3397: /* Careful transposed matrix */
1.266 brouard 3398: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3399: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3400: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3401: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3402: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3403: /* if((int)age == 70){ */
3404: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3405: /* for(i=1; i<=nlstate+ndeath; i++) { */
3406: /* printf("%d pmmij ",i); */
3407: /* for(j=1;j<=nlstate+ndeath;j++) { */
3408: /* printf("%f ",pmmij[i][j]); */
3409: /* } */
3410: /* printf(" oldm "); */
3411: /* for(j=1;j<=nlstate+ndeath;j++) { */
3412: /* printf("%f ",oldm[i][j]); */
3413: /* } */
3414: /* printf("\n"); */
3415: /* } */
3416: /* } */
3417: savm=oldm;
3418: oldm=newm;
3419: }
3420: for(i=1; i<=nlstate+ndeath; i++)
3421: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3422: po[i][j][h]=newm[i][j];
1.268 brouard 3423: /* if(h==nhstepm) */
3424: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3425: }
1.268 brouard 3426: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3427: } /* end h */
1.268 brouard 3428: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3429: return po;
3430: }
3431:
3432:
1.162 brouard 3433: #ifdef NLOPT
3434: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3435: double fret;
3436: double *xt;
3437: int j;
3438: myfunc_data *d2 = (myfunc_data *) pd;
3439: /* xt = (p1-1); */
3440: xt=vector(1,n);
3441: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3442:
3443: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3444: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3445: printf("Function = %.12lf ",fret);
3446: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3447: printf("\n");
3448: free_vector(xt,1,n);
3449: return fret;
3450: }
3451: #endif
1.126 brouard 3452:
3453: /*************** log-likelihood *************/
3454: double func( double *x)
3455: {
1.226 brouard 3456: int i, ii, j, k, mi, d, kk;
3457: int ioffset=0;
3458: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3459: double **out;
3460: double lli; /* Individual log likelihood */
3461: int s1, s2;
1.228 brouard 3462: 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 3463: double bbh, survp;
3464: long ipmx;
3465: double agexact;
3466: /*extern weight */
3467: /* We are differentiating ll according to initial status */
3468: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3469: /*for(i=1;i<imx;i++)
3470: printf(" %d\n",s[4][i]);
3471: */
1.162 brouard 3472:
1.226 brouard 3473: ++countcallfunc;
1.162 brouard 3474:
1.226 brouard 3475: cov[1]=1.;
1.126 brouard 3476:
1.226 brouard 3477: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3478: ioffset=0;
1.226 brouard 3479: if(mle==1){
3480: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3481: /* Computes the values of the ncovmodel covariates of the model
3482: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3483: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3484: to be observed in j being in i according to the model.
3485: */
1.243 brouard 3486: ioffset=2+nagesqr ;
1.233 brouard 3487: /* Fixed */
1.234 brouard 3488: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3489: 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)*/
3490: }
1.226 brouard 3491: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3492: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3493: has been calculated etc */
3494: /* For an individual i, wav[i] gives the number of effective waves */
3495: /* We compute the contribution to Likelihood of each effective transition
3496: mw[mi][i] is real wave of the mi th effectve wave */
3497: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3498: s2=s[mw[mi+1][i]][i];
3499: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3500: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3501: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3502: */
3503: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3504: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3505: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3506: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3507: }
3508: for (ii=1;ii<=nlstate+ndeath;ii++)
3509: for (j=1;j<=nlstate+ndeath;j++){
3510: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3511: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3512: }
3513: for(d=0; d<dh[mi][i]; d++){
3514: newm=savm;
3515: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3516: cov[2]=agexact;
3517: if(nagesqr==1)
3518: cov[3]= agexact*agexact; /* Should be changed here */
3519: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3520: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3521: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3522: else
3523: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3524: }
3525: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3526: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3527: savm=oldm;
3528: oldm=newm;
3529: } /* end mult */
3530:
3531: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3532: /* But now since version 0.9 we anticipate for bias at large stepm.
3533: * If stepm is larger than one month (smallest stepm) and if the exact delay
3534: * (in months) between two waves is not a multiple of stepm, we rounded to
3535: * the nearest (and in case of equal distance, to the lowest) interval but now
3536: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3537: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3538: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3539: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3540: * -stepm/2 to stepm/2 .
3541: * For stepm=1 the results are the same as for previous versions of Imach.
3542: * For stepm > 1 the results are less biased than in previous versions.
3543: */
1.234 brouard 3544: s1=s[mw[mi][i]][i];
3545: s2=s[mw[mi+1][i]][i];
3546: bbh=(double)bh[mi][i]/(double)stepm;
3547: /* bias bh is positive if real duration
3548: * is higher than the multiple of stepm and negative otherwise.
3549: */
3550: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3551: if( s2 > nlstate){
3552: /* i.e. if s2 is a death state and if the date of death is known
3553: then the contribution to the likelihood is the probability to
3554: die between last step unit time and current step unit time,
3555: which is also equal to probability to die before dh
3556: minus probability to die before dh-stepm .
3557: In version up to 0.92 likelihood was computed
3558: as if date of death was unknown. Death was treated as any other
3559: health state: the date of the interview describes the actual state
3560: and not the date of a change in health state. The former idea was
3561: to consider that at each interview the state was recorded
3562: (healthy, disable or death) and IMaCh was corrected; but when we
3563: introduced the exact date of death then we should have modified
3564: the contribution of an exact death to the likelihood. This new
3565: contribution is smaller and very dependent of the step unit
3566: stepm. It is no more the probability to die between last interview
3567: and month of death but the probability to survive from last
3568: interview up to one month before death multiplied by the
3569: probability to die within a month. Thanks to Chris
3570: Jackson for correcting this bug. Former versions increased
3571: mortality artificially. The bad side is that we add another loop
3572: which slows down the processing. The difference can be up to 10%
3573: lower mortality.
3574: */
3575: /* If, at the beginning of the maximization mostly, the
3576: cumulative probability or probability to be dead is
3577: constant (ie = 1) over time d, the difference is equal to
3578: 0. out[s1][3] = savm[s1][3]: probability, being at state
3579: s1 at precedent wave, to be dead a month before current
3580: wave is equal to probability, being at state s1 at
3581: precedent wave, to be dead at mont of the current
3582: wave. Then the observed probability (that this person died)
3583: is null according to current estimated parameter. In fact,
3584: it should be very low but not zero otherwise the log go to
3585: infinity.
3586: */
1.183 brouard 3587: /* #ifdef INFINITYORIGINAL */
3588: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3589: /* #else */
3590: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3591: /* lli=log(mytinydouble); */
3592: /* else */
3593: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3594: /* #endif */
1.226 brouard 3595: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3596:
1.226 brouard 3597: } else if ( s2==-1 ) { /* alive */
3598: for (j=1,survp=0. ; j<=nlstate; j++)
3599: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3600: /*survp += out[s1][j]; */
3601: lli= log(survp);
3602: }
3603: else if (s2==-4) {
3604: for (j=3,survp=0. ; j<=nlstate; j++)
3605: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3606: lli= log(survp);
3607: }
3608: else if (s2==-5) {
3609: for (j=1,survp=0. ; j<=2; j++)
3610: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3611: lli= log(survp);
3612: }
3613: else{
3614: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3615: /* 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 */
3616: }
3617: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3618: /*if(lli ==000.0)*/
3619: /*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); */
3620: ipmx +=1;
3621: sw += weight[i];
3622: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3623: /* if (lli < log(mytinydouble)){ */
3624: /* 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); */
3625: /* 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]); */
3626: /* } */
3627: } /* end of wave */
3628: } /* end of individual */
3629: } else if(mle==2){
3630: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3631: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3632: for(mi=1; mi<= wav[i]-1; mi++){
3633: for (ii=1;ii<=nlstate+ndeath;ii++)
3634: for (j=1;j<=nlstate+ndeath;j++){
3635: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3636: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3637: }
3638: for(d=0; d<=dh[mi][i]; d++){
3639: newm=savm;
3640: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3641: cov[2]=agexact;
3642: if(nagesqr==1)
3643: cov[3]= agexact*agexact;
3644: for (kk=1; kk<=cptcovage;kk++) {
3645: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3646: }
3647: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3648: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3649: savm=oldm;
3650: oldm=newm;
3651: } /* end mult */
3652:
3653: s1=s[mw[mi][i]][i];
3654: s2=s[mw[mi+1][i]][i];
3655: bbh=(double)bh[mi][i]/(double)stepm;
3656: 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 */
3657: ipmx +=1;
3658: sw += weight[i];
3659: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3660: } /* end of wave */
3661: } /* end of individual */
3662: } else if(mle==3){ /* exponential inter-extrapolation */
3663: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3664: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3665: for(mi=1; mi<= wav[i]-1; mi++){
3666: for (ii=1;ii<=nlstate+ndeath;ii++)
3667: for (j=1;j<=nlstate+ndeath;j++){
3668: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3669: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3670: }
3671: for(d=0; d<dh[mi][i]; d++){
3672: newm=savm;
3673: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3674: cov[2]=agexact;
3675: if(nagesqr==1)
3676: cov[3]= agexact*agexact;
3677: for (kk=1; kk<=cptcovage;kk++) {
3678: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3679: }
3680: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3681: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3682: savm=oldm;
3683: oldm=newm;
3684: } /* end mult */
3685:
3686: s1=s[mw[mi][i]][i];
3687: s2=s[mw[mi+1][i]][i];
3688: bbh=(double)bh[mi][i]/(double)stepm;
3689: 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 */
3690: ipmx +=1;
3691: sw += weight[i];
3692: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3693: } /* end of wave */
3694: } /* end of individual */
3695: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3696: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3697: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3698: for(mi=1; mi<= wav[i]-1; mi++){
3699: for (ii=1;ii<=nlstate+ndeath;ii++)
3700: for (j=1;j<=nlstate+ndeath;j++){
3701: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3702: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3703: }
3704: for(d=0; d<dh[mi][i]; d++){
3705: newm=savm;
3706: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3707: cov[2]=agexact;
3708: if(nagesqr==1)
3709: cov[3]= agexact*agexact;
3710: for (kk=1; kk<=cptcovage;kk++) {
3711: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3712: }
1.126 brouard 3713:
1.226 brouard 3714: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3715: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3716: savm=oldm;
3717: oldm=newm;
3718: } /* end mult */
3719:
3720: s1=s[mw[mi][i]][i];
3721: s2=s[mw[mi+1][i]][i];
3722: if( s2 > nlstate){
3723: lli=log(out[s1][s2] - savm[s1][s2]);
3724: } else if ( s2==-1 ) { /* alive */
3725: for (j=1,survp=0. ; j<=nlstate; j++)
3726: survp += out[s1][j];
3727: lli= log(survp);
3728: }else{
3729: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3730: }
3731: ipmx +=1;
3732: sw += weight[i];
3733: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3734: /* 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 3735: } /* end of wave */
3736: } /* end of individual */
3737: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3738: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3739: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3740: for(mi=1; mi<= wav[i]-1; mi++){
3741: for (ii=1;ii<=nlstate+ndeath;ii++)
3742: for (j=1;j<=nlstate+ndeath;j++){
3743: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3744: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3745: }
3746: for(d=0; d<dh[mi][i]; d++){
3747: newm=savm;
3748: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3749: cov[2]=agexact;
3750: if(nagesqr==1)
3751: cov[3]= agexact*agexact;
3752: for (kk=1; kk<=cptcovage;kk++) {
3753: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3754: }
1.126 brouard 3755:
1.226 brouard 3756: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3757: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3758: savm=oldm;
3759: oldm=newm;
3760: } /* end mult */
3761:
3762: s1=s[mw[mi][i]][i];
3763: s2=s[mw[mi+1][i]][i];
3764: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3765: ipmx +=1;
3766: sw += weight[i];
3767: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3768: /*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]);*/
3769: } /* end of wave */
3770: } /* end of individual */
3771: } /* End of if */
3772: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3773: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3774: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3775: return -l;
1.126 brouard 3776: }
3777:
3778: /*************** log-likelihood *************/
3779: double funcone( double *x)
3780: {
1.228 brouard 3781: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3782: int i, ii, j, k, mi, d, kk;
1.228 brouard 3783: int ioffset=0;
1.131 brouard 3784: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3785: double **out;
3786: double lli; /* Individual log likelihood */
3787: double llt;
3788: int s1, s2;
1.228 brouard 3789: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3790:
1.126 brouard 3791: double bbh, survp;
1.187 brouard 3792: double agexact;
1.214 brouard 3793: double agebegin, ageend;
1.126 brouard 3794: /*extern weight */
3795: /* We are differentiating ll according to initial status */
3796: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3797: /*for(i=1;i<imx;i++)
3798: printf(" %d\n",s[4][i]);
3799: */
3800: cov[1]=1.;
3801:
3802: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3803: ioffset=0;
3804: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3805: /* ioffset=2+nagesqr+cptcovage; */
3806: ioffset=2+nagesqr;
1.232 brouard 3807: /* Fixed */
1.224 brouard 3808: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3809: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3810: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3811: 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)*/
3812: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3813: /* cov[2+6]=covar[Tvar[6]][i]; */
3814: /* cov[2+6]=covar[2][i]; V2 */
3815: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3816: /* cov[2+7]=covar[Tvar[7]][i]; */
3817: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3818: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3819: /* cov[2+9]=covar[Tvar[9]][i]; */
3820: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3821: }
1.232 brouard 3822: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3823: /* 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?)*\/ */
3824: /* } */
1.231 brouard 3825: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3826: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3827: /* } */
1.225 brouard 3828:
1.233 brouard 3829:
3830: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3831: /* Wave varying (but not age varying) */
3832: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3833: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3834: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3835: }
1.232 brouard 3836: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3837: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3838: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3839: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3840: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3841: /* 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 3842: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3843: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3844: /* /\* 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]); *\/ */
3845: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3846: /* } */
1.126 brouard 3847: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3848: for (j=1;j<=nlstate+ndeath;j++){
3849: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3850: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3851: }
1.214 brouard 3852:
3853: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3854: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3855: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3856: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3857: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3858: and mw[mi+1][i]. dh depends on stepm.*/
3859: newm=savm;
1.247 brouard 3860: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3861: cov[2]=agexact;
3862: if(nagesqr==1)
3863: cov[3]= agexact*agexact;
3864: for (kk=1; kk<=cptcovage;kk++) {
3865: if(!FixedV[Tvar[Tage[kk]]])
3866: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3867: else
3868: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3869: }
3870: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3871: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3872: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3873: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3874: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3875: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3876: savm=oldm;
3877: oldm=newm;
1.126 brouard 3878: } /* end mult */
3879:
3880: s1=s[mw[mi][i]][i];
3881: s2=s[mw[mi+1][i]][i];
1.217 brouard 3882: /* if(s2==-1){ */
1.268 brouard 3883: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3884: /* /\* exit(1); *\/ */
3885: /* } */
1.126 brouard 3886: bbh=(double)bh[mi][i]/(double)stepm;
3887: /* bias is positive if real duration
3888: * is higher than the multiple of stepm and negative otherwise.
3889: */
3890: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3891: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3892: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3893: for (j=1,survp=0. ; j<=nlstate; j++)
3894: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3895: lli= log(survp);
1.126 brouard 3896: }else if (mle==1){
1.242 brouard 3897: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3898: } else if(mle==2){
1.242 brouard 3899: 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 3900: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3901: 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 3902: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3903: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3904: } else{ /* mle=0 back to 1 */
1.242 brouard 3905: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3906: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3907: } /* End of if */
3908: ipmx +=1;
3909: sw += weight[i];
3910: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3911: /*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 3912: if(globpr){
1.246 brouard 3913: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3914: %11.6f %11.6f %11.6f ", \
1.242 brouard 3915: 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 3916: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3917: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3918: llt +=ll[k]*gipmx/gsw;
3919: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3920: }
3921: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3922: }
1.232 brouard 3923: } /* end of wave */
3924: } /* end of individual */
3925: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3926: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3927: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3928: if(globpr==0){ /* First time we count the contributions and weights */
3929: gipmx=ipmx;
3930: gsw=sw;
3931: }
3932: return -l;
1.126 brouard 3933: }
3934:
3935:
3936: /*************** function likelione ***********/
1.292 brouard 3937: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3938: {
3939: /* This routine should help understanding what is done with
3940: the selection of individuals/waves and
3941: to check the exact contribution to the likelihood.
3942: Plotting could be done.
3943: */
3944: int k;
3945:
3946: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3947: strcpy(fileresilk,"ILK_");
1.202 brouard 3948: strcat(fileresilk,fileresu);
1.126 brouard 3949: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3950: printf("Problem with resultfile: %s\n", fileresilk);
3951: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3952: }
1.214 brouard 3953: 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");
3954: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3955: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3956: for(k=1; k<=nlstate; k++)
3957: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3958: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3959: }
3960:
1.292 brouard 3961: *fretone=(*func)(p);
1.126 brouard 3962: if(*globpri !=0){
3963: fclose(ficresilk);
1.205 brouard 3964: if (mle ==0)
3965: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3966: else if(mle >=1)
3967: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3968: 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 3969: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3970:
3971: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3972: 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 3973: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3974: }
1.207 brouard 3975: 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 3976: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3977: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3978: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3979: fflush(fichtm);
1.205 brouard 3980: }
1.126 brouard 3981: return;
3982: }
3983:
3984:
3985: /*********** Maximum Likelihood Estimation ***************/
3986:
3987: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3988: {
1.165 brouard 3989: int i,j, iter=0;
1.126 brouard 3990: double **xi;
3991: double fret;
3992: double fretone; /* Only one call to likelihood */
3993: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3994:
3995: #ifdef NLOPT
3996: int creturn;
3997: nlopt_opt opt;
3998: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3999: double *lb;
4000: double minf; /* the minimum objective value, upon return */
4001: double * p1; /* Shifted parameters from 0 instead of 1 */
4002: myfunc_data dinst, *d = &dinst;
4003: #endif
4004:
4005:
1.126 brouard 4006: xi=matrix(1,npar,1,npar);
4007: for (i=1;i<=npar;i++)
4008: for (j=1;j<=npar;j++)
4009: xi[i][j]=(i==j ? 1.0 : 0.0);
4010: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4011: strcpy(filerespow,"POW_");
1.126 brouard 4012: strcat(filerespow,fileres);
4013: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4014: printf("Problem with resultfile: %s\n", filerespow);
4015: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4016: }
4017: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4018: for (i=1;i<=nlstate;i++)
4019: for(j=1;j<=nlstate+ndeath;j++)
4020: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4021: fprintf(ficrespow,"\n");
1.162 brouard 4022: #ifdef POWELL
1.126 brouard 4023: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4024: #endif
1.126 brouard 4025:
1.162 brouard 4026: #ifdef NLOPT
4027: #ifdef NEWUOA
4028: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4029: #else
4030: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4031: #endif
4032: lb=vector(0,npar-1);
4033: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4034: nlopt_set_lower_bounds(opt, lb);
4035: nlopt_set_initial_step1(opt, 0.1);
4036:
4037: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4038: d->function = func;
4039: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4040: nlopt_set_min_objective(opt, myfunc, d);
4041: nlopt_set_xtol_rel(opt, ftol);
4042: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4043: printf("nlopt failed! %d\n",creturn);
4044: }
4045: else {
4046: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4047: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4048: iter=1; /* not equal */
4049: }
4050: nlopt_destroy(opt);
4051: #endif
1.126 brouard 4052: free_matrix(xi,1,npar,1,npar);
4053: fclose(ficrespow);
1.203 brouard 4054: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4055: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4056: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4057:
4058: }
4059:
4060: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4061: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4062: {
4063: double **a,**y,*x,pd;
1.203 brouard 4064: /* double **hess; */
1.164 brouard 4065: int i, j;
1.126 brouard 4066: int *indx;
4067:
4068: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4069: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4070: void lubksb(double **a, int npar, int *indx, double b[]) ;
4071: void ludcmp(double **a, int npar, int *indx, double *d) ;
4072: double gompertz(double p[]);
1.203 brouard 4073: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4074:
4075: printf("\nCalculation of the hessian matrix. Wait...\n");
4076: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4077: for (i=1;i<=npar;i++){
1.203 brouard 4078: printf("%d-",i);fflush(stdout);
4079: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4080:
4081: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4082:
4083: /* printf(" %f ",p[i]);
4084: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4085: }
4086:
4087: for (i=1;i<=npar;i++) {
4088: for (j=1;j<=npar;j++) {
4089: if (j>i) {
1.203 brouard 4090: printf(".%d-%d",i,j);fflush(stdout);
4091: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4092: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4093:
4094: hess[j][i]=hess[i][j];
4095: /*printf(" %lf ",hess[i][j]);*/
4096: }
4097: }
4098: }
4099: printf("\n");
4100: fprintf(ficlog,"\n");
4101:
4102: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4103: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4104:
4105: a=matrix(1,npar,1,npar);
4106: y=matrix(1,npar,1,npar);
4107: x=vector(1,npar);
4108: indx=ivector(1,npar);
4109: for (i=1;i<=npar;i++)
4110: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4111: ludcmp(a,npar,indx,&pd);
4112:
4113: for (j=1;j<=npar;j++) {
4114: for (i=1;i<=npar;i++) x[i]=0;
4115: x[j]=1;
4116: lubksb(a,npar,indx,x);
4117: for (i=1;i<=npar;i++){
4118: matcov[i][j]=x[i];
4119: }
4120: }
4121:
4122: printf("\n#Hessian matrix#\n");
4123: fprintf(ficlog,"\n#Hessian matrix#\n");
4124: for (i=1;i<=npar;i++) {
4125: for (j=1;j<=npar;j++) {
1.203 brouard 4126: printf("%.6e ",hess[i][j]);
4127: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4128: }
4129: printf("\n");
4130: fprintf(ficlog,"\n");
4131: }
4132:
1.203 brouard 4133: /* printf("\n#Covariance matrix#\n"); */
4134: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4135: /* for (i=1;i<=npar;i++) { */
4136: /* for (j=1;j<=npar;j++) { */
4137: /* printf("%.6e ",matcov[i][j]); */
4138: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4139: /* } */
4140: /* printf("\n"); */
4141: /* fprintf(ficlog,"\n"); */
4142: /* } */
4143:
1.126 brouard 4144: /* Recompute Inverse */
1.203 brouard 4145: /* for (i=1;i<=npar;i++) */
4146: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4147: /* ludcmp(a,npar,indx,&pd); */
4148:
4149: /* printf("\n#Hessian matrix recomputed#\n"); */
4150:
4151: /* for (j=1;j<=npar;j++) { */
4152: /* for (i=1;i<=npar;i++) x[i]=0; */
4153: /* x[j]=1; */
4154: /* lubksb(a,npar,indx,x); */
4155: /* for (i=1;i<=npar;i++){ */
4156: /* y[i][j]=x[i]; */
4157: /* printf("%.3e ",y[i][j]); */
4158: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4159: /* } */
4160: /* printf("\n"); */
4161: /* fprintf(ficlog,"\n"); */
4162: /* } */
4163:
4164: /* Verifying the inverse matrix */
4165: #ifdef DEBUGHESS
4166: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4167:
1.203 brouard 4168: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4169: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4170:
4171: for (j=1;j<=npar;j++) {
4172: for (i=1;i<=npar;i++){
1.203 brouard 4173: printf("%.2f ",y[i][j]);
4174: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4175: }
4176: printf("\n");
4177: fprintf(ficlog,"\n");
4178: }
1.203 brouard 4179: #endif
1.126 brouard 4180:
4181: free_matrix(a,1,npar,1,npar);
4182: free_matrix(y,1,npar,1,npar);
4183: free_vector(x,1,npar);
4184: free_ivector(indx,1,npar);
1.203 brouard 4185: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4186:
4187:
4188: }
4189:
4190: /*************** hessian matrix ****************/
4191: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4192: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4193: int i;
4194: int l=1, lmax=20;
1.203 brouard 4195: double k1,k2, res, fx;
1.132 brouard 4196: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4197: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4198: int k=0,kmax=10;
4199: double l1;
4200:
4201: fx=func(x);
4202: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4203: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4204: l1=pow(10,l);
4205: delts=delt;
4206: for(k=1 ; k <kmax; k=k+1){
4207: delt = delta*(l1*k);
4208: p2[theta]=x[theta] +delt;
1.145 brouard 4209: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4210: p2[theta]=x[theta]-delt;
4211: k2=func(p2)-fx;
4212: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4213: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4214:
1.203 brouard 4215: #ifdef DEBUGHESSII
1.126 brouard 4216: 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);
4217: 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);
4218: #endif
4219: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4220: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4221: k=kmax;
4222: }
4223: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4224: k=kmax; l=lmax*10;
1.126 brouard 4225: }
4226: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4227: delts=delt;
4228: }
1.203 brouard 4229: } /* End loop k */
1.126 brouard 4230: }
4231: delti[theta]=delts;
4232: return res;
4233:
4234: }
4235:
1.203 brouard 4236: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4237: {
4238: int i;
1.164 brouard 4239: int l=1, lmax=20;
1.126 brouard 4240: double k1,k2,k3,k4,res,fx;
1.132 brouard 4241: double p2[MAXPARM+1];
1.203 brouard 4242: int k, kmax=1;
4243: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4244:
4245: int firstime=0;
1.203 brouard 4246:
1.126 brouard 4247: fx=func(x);
1.203 brouard 4248: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4249: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4250: p2[thetai]=x[thetai]+delti[thetai]*k;
4251: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4252: k1=func(p2)-fx;
4253:
1.203 brouard 4254: p2[thetai]=x[thetai]+delti[thetai]*k;
4255: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4256: k2=func(p2)-fx;
4257:
1.203 brouard 4258: p2[thetai]=x[thetai]-delti[thetai]*k;
4259: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4260: k3=func(p2)-fx;
4261:
1.203 brouard 4262: p2[thetai]=x[thetai]-delti[thetai]*k;
4263: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4264: k4=func(p2)-fx;
1.203 brouard 4265: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4266: if(k1*k2*k3*k4 <0.){
1.208 brouard 4267: firstime=1;
1.203 brouard 4268: kmax=kmax+10;
1.208 brouard 4269: }
4270: if(kmax >=10 || firstime ==1){
1.246 brouard 4271: 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);
4272: 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 4273: 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);
4274: 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);
4275: }
4276: #ifdef DEBUGHESSIJ
4277: v1=hess[thetai][thetai];
4278: v2=hess[thetaj][thetaj];
4279: cv12=res;
4280: /* Computing eigen value of Hessian matrix */
4281: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4282: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4283: if ((lc2 <0) || (lc1 <0) ){
4284: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4285: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4286: 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);
4287: 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);
4288: }
1.126 brouard 4289: #endif
4290: }
4291: return res;
4292: }
4293:
1.203 brouard 4294: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4295: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4296: /* { */
4297: /* int i; */
4298: /* int l=1, lmax=20; */
4299: /* double k1,k2,k3,k4,res,fx; */
4300: /* double p2[MAXPARM+1]; */
4301: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4302: /* int k=0,kmax=10; */
4303: /* double l1; */
4304:
4305: /* fx=func(x); */
4306: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4307: /* l1=pow(10,l); */
4308: /* delts=delt; */
4309: /* for(k=1 ; k <kmax; k=k+1){ */
4310: /* delt = delti*(l1*k); */
4311: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4312: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4313: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4314: /* k1=func(p2)-fx; */
4315:
4316: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4317: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4318: /* k2=func(p2)-fx; */
4319:
4320: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4321: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4322: /* k3=func(p2)-fx; */
4323:
4324: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4325: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4326: /* k4=func(p2)-fx; */
4327: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4328: /* #ifdef DEBUGHESSIJ */
4329: /* 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); */
4330: /* 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); */
4331: /* #endif */
4332: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4333: /* k=kmax; */
4334: /* } */
4335: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4336: /* k=kmax; l=lmax*10; */
4337: /* } */
4338: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4339: /* delts=delt; */
4340: /* } */
4341: /* } /\* End loop k *\/ */
4342: /* } */
4343: /* delti[theta]=delts; */
4344: /* return res; */
4345: /* } */
4346:
4347:
1.126 brouard 4348: /************** Inverse of matrix **************/
4349: void ludcmp(double **a, int n, int *indx, double *d)
4350: {
4351: int i,imax,j,k;
4352: double big,dum,sum,temp;
4353: double *vv;
4354:
4355: vv=vector(1,n);
4356: *d=1.0;
4357: for (i=1;i<=n;i++) {
4358: big=0.0;
4359: for (j=1;j<=n;j++)
4360: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4361: if (big == 0.0){
4362: printf(" Singular Hessian matrix at row %d:\n",i);
4363: for (j=1;j<=n;j++) {
4364: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4365: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4366: }
4367: fflush(ficlog);
4368: fclose(ficlog);
4369: nrerror("Singular matrix in routine ludcmp");
4370: }
1.126 brouard 4371: vv[i]=1.0/big;
4372: }
4373: for (j=1;j<=n;j++) {
4374: for (i=1;i<j;i++) {
4375: sum=a[i][j];
4376: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4377: a[i][j]=sum;
4378: }
4379: big=0.0;
4380: for (i=j;i<=n;i++) {
4381: sum=a[i][j];
4382: for (k=1;k<j;k++)
4383: sum -= a[i][k]*a[k][j];
4384: a[i][j]=sum;
4385: if ( (dum=vv[i]*fabs(sum)) >= big) {
4386: big=dum;
4387: imax=i;
4388: }
4389: }
4390: if (j != imax) {
4391: for (k=1;k<=n;k++) {
4392: dum=a[imax][k];
4393: a[imax][k]=a[j][k];
4394: a[j][k]=dum;
4395: }
4396: *d = -(*d);
4397: vv[imax]=vv[j];
4398: }
4399: indx[j]=imax;
4400: if (a[j][j] == 0.0) a[j][j]=TINY;
4401: if (j != n) {
4402: dum=1.0/(a[j][j]);
4403: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4404: }
4405: }
4406: free_vector(vv,1,n); /* Doesn't work */
4407: ;
4408: }
4409:
4410: void lubksb(double **a, int n, int *indx, double b[])
4411: {
4412: int i,ii=0,ip,j;
4413: double sum;
4414:
4415: for (i=1;i<=n;i++) {
4416: ip=indx[i];
4417: sum=b[ip];
4418: b[ip]=b[i];
4419: if (ii)
4420: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4421: else if (sum) ii=i;
4422: b[i]=sum;
4423: }
4424: for (i=n;i>=1;i--) {
4425: sum=b[i];
4426: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4427: b[i]=sum/a[i][i];
4428: }
4429: }
4430:
4431: void pstamp(FILE *fichier)
4432: {
1.196 brouard 4433: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4434: }
4435:
1.297 brouard 4436: void date2dmy(double date,double *day, double *month, double *year){
4437: double yp=0., yp1=0., yp2=0.;
4438:
4439: yp1=modf(date,&yp);/* extracts integral of date in yp and
4440: fractional in yp1 */
4441: *year=yp;
4442: yp2=modf((yp1*12),&yp);
4443: *month=yp;
4444: yp1=modf((yp2*30.5),&yp);
4445: *day=yp;
4446: if(*day==0) *day=1;
4447: if(*month==0) *month=1;
4448: }
4449:
1.253 brouard 4450:
4451:
1.126 brouard 4452: /************ Frequencies ********************/
1.251 brouard 4453: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4454: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4455: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4456: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4457:
1.265 brouard 4458: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4459: int iind=0, iage=0;
4460: int mi; /* Effective wave */
4461: int first;
4462: double ***freq; /* Frequencies */
1.268 brouard 4463: 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 */
4464: 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 4465: double *meanq, *stdq, *idq;
1.226 brouard 4466: double **meanqt;
4467: double *pp, **prop, *posprop, *pospropt;
4468: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4469: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4470: double agebegin, ageend;
4471:
4472: pp=vector(1,nlstate);
1.251 brouard 4473: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4474: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4475: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4476: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4477: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4478: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4479: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4480: meanqt=matrix(1,lastpass,1,nqtveff);
4481: strcpy(fileresp,"P_");
4482: strcat(fileresp,fileresu);
4483: /*strcat(fileresphtm,fileresu);*/
4484: if((ficresp=fopen(fileresp,"w"))==NULL) {
4485: printf("Problem with prevalence resultfile: %s\n", fileresp);
4486: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4487: exit(0);
4488: }
1.240 brouard 4489:
1.226 brouard 4490: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4491: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4492: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4493: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4494: fflush(ficlog);
4495: exit(70);
4496: }
4497: else{
4498: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4499: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4500: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4501: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4502: }
1.237 brouard 4503: 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 4504:
1.226 brouard 4505: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4506: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4507: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4508: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4509: fflush(ficlog);
4510: exit(70);
1.240 brouard 4511: } else{
1.226 brouard 4512: 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 4513: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4514: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4515: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4516: }
1.240 brouard 4517: 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);
4518:
1.253 brouard 4519: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4520: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4521: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4522: j1=0;
1.126 brouard 4523:
1.227 brouard 4524: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4525: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4526: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4527:
4528:
1.226 brouard 4529: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4530: reference=low_education V1=0,V2=0
4531: med_educ V1=1 V2=0,
4532: high_educ V1=0 V2=1
4533: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4534: */
1.249 brouard 4535: dateintsum=0;
4536: k2cpt=0;
4537:
1.253 brouard 4538: if(cptcoveff == 0 )
1.265 brouard 4539: nl=1; /* Constant and age model only */
1.253 brouard 4540: else
4541: nl=2;
1.265 brouard 4542:
4543: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4544: /* Loop on nj=1 or 2 if dummy covariates j!=0
4545: * Loop on j1(1 to 2**cptcoveff) covariate combination
4546: * freq[s1][s2][iage] =0.
4547: * Loop on iind
4548: * ++freq[s1][s2][iage] weighted
4549: * end iind
4550: * if covariate and j!0
4551: * headers Variable on one line
4552: * endif cov j!=0
4553: * header of frequency table by age
4554: * Loop on age
4555: * pp[s1]+=freq[s1][s2][iage] weighted
4556: * pos+=freq[s1][s2][iage] weighted
4557: * Loop on s1 initial state
4558: * fprintf(ficresp
4559: * end s1
4560: * end age
4561: * if j!=0 computes starting values
4562: * end compute starting values
4563: * end j1
4564: * end nl
4565: */
1.253 brouard 4566: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4567: if(nj==1)
4568: j=0; /* First pass for the constant */
1.265 brouard 4569: else{
1.253 brouard 4570: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4571: }
1.251 brouard 4572: first=1;
1.265 brouard 4573: 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 4574: posproptt=0.;
4575: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4576: scanf("%d", i);*/
4577: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4578: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4579: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4580: freq[i][s2][m]=0;
1.251 brouard 4581:
4582: for (i=1; i<=nlstate; i++) {
1.240 brouard 4583: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4584: prop[i][m]=0;
4585: posprop[i]=0;
4586: pospropt[i]=0;
4587: }
1.283 brouard 4588: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4589: idq[z1]=0.;
4590: meanq[z1]=0.;
4591: stdq[z1]=0.;
1.283 brouard 4592: }
4593: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4594: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4595: /* meanqt[m][z1]=0.; */
4596: /* } */
4597: /* } */
1.251 brouard 4598: /* dateintsum=0; */
4599: /* k2cpt=0; */
4600:
1.265 brouard 4601: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4602: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4603: bool=1;
4604: if(j !=0){
4605: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4606: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4607: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4608: /* if(Tvaraff[z1] ==-20){ */
4609: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4610: /* }else if(Tvaraff[z1] ==-10){ */
4611: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4612: /* }else */
4613: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4614: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4615: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4616: /* 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",
4617: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4618: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4619: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4620: } /* Onlyf fixed */
4621: } /* end z1 */
4622: } /* cptcovn > 0 */
4623: } /* end any */
4624: }/* end j==0 */
1.265 brouard 4625: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4626: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4627: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4628: m=mw[mi][iind];
4629: if(j!=0){
4630: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4631: for (z1=1; z1<=cptcoveff; z1++) {
4632: if( Fixed[Tmodelind[z1]]==1){
4633: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4634: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4635: value is -1, we don't select. It differs from the
4636: constant and age model which counts them. */
4637: bool=0; /* not selected */
4638: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4639: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4640: bool=0;
4641: }
4642: }
4643: }
4644: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4645: } /* end j==0 */
4646: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4647: if(bool==1){ /*Selected */
1.251 brouard 4648: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4649: and mw[mi+1][iind]. dh depends on stepm. */
4650: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4651: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4652: if(m >=firstpass && m <=lastpass){
4653: k2=anint[m][iind]+(mint[m][iind]/12.);
4654: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4655: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4656: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4657: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4658: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4659: if (m<lastpass) {
4660: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4661: /* 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]); */
4662: if(s[m][iind]==-1)
4663: 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.));
4664: 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 4665: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4666: idq[z1]=idq[z1]+weight[iind];
4667: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4668: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4669: }
1.251 brouard 4670: /* if((int)agev[m][iind] == 55) */
4671: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4672: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4673: 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 4674: }
1.251 brouard 4675: } /* end if between passes */
4676: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4677: dateintsum=dateintsum+k2; /* on all covariates ?*/
4678: k2cpt++;
4679: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4680: }
1.251 brouard 4681: }else{
4682: bool=1;
4683: }/* end bool 2 */
4684: } /* end m */
1.284 brouard 4685: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4686: /* idq[z1]=idq[z1]+weight[iind]; */
4687: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4688: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4689: /* } */
1.251 brouard 4690: } /* end bool */
4691: } /* end iind = 1 to imx */
4692: /* prop[s][age] is feeded for any initial and valid live state as well as
4693: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4694:
4695:
4696: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4697: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4698: pstamp(ficresp);
1.251 brouard 4699: if (cptcoveff>0 && j!=0){
1.265 brouard 4700: pstamp(ficresp);
1.251 brouard 4701: printf( "\n#********** Variable ");
4702: fprintf(ficresp, "\n#********** Variable ");
4703: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4704: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4705: fprintf(ficlog, "\n#********** Variable ");
4706: for (z1=1; z1<=cptcoveff; z1++){
4707: if(!FixedV[Tvaraff[z1]]){
4708: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4709: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4710: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4711: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4712: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4713: }else{
1.251 brouard 4714: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4715: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4716: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4717: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4718: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4719: }
4720: }
4721: printf( "**********\n#");
4722: fprintf(ficresp, "**********\n#");
4723: fprintf(ficresphtm, "**********</h3>\n");
4724: fprintf(ficresphtmfr, "**********</h3>\n");
4725: fprintf(ficlog, "**********\n");
4726: }
1.284 brouard 4727: /*
4728: Printing means of quantitative variables if any
4729: */
4730: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4731: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4732: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4733: if(weightopt==1){
4734: printf(" Weighted mean and standard deviation of");
4735: fprintf(ficlog," Weighted mean and standard deviation of");
4736: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4737: }
1.285 brouard 4738: 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]));
4739: 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]));
4740: 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 4741: }
4742: /* for (z1=1; z1<= nqtveff; z1++) { */
4743: /* for(m=1;m<=lastpass;m++){ */
4744: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4745: /* } */
4746: /* } */
1.283 brouard 4747:
1.251 brouard 4748: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4749: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4750: fprintf(ficresp, " Age");
4751: 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 4752: for(i=1; i<=nlstate;i++) {
1.265 brouard 4753: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4754: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4755: }
1.265 brouard 4756: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4757: fprintf(ficresphtm, "\n");
4758:
4759: /* Header of frequency table by age */
4760: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4761: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4762: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4763: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4764: if(s2!=0 && m!=0)
4765: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4766: }
1.226 brouard 4767: }
1.251 brouard 4768: fprintf(ficresphtmfr, "\n");
4769:
4770: /* For each age */
4771: for(iage=iagemin; iage <= iagemax+3; iage++){
4772: fprintf(ficresphtm,"<tr>");
4773: if(iage==iagemax+1){
4774: fprintf(ficlog,"1");
4775: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4776: }else if(iage==iagemax+2){
4777: fprintf(ficlog,"0");
4778: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4779: }else if(iage==iagemax+3){
4780: fprintf(ficlog,"Total");
4781: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4782: }else{
1.240 brouard 4783: if(first==1){
1.251 brouard 4784: first=0;
4785: printf("See log file for details...\n");
4786: }
4787: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4788: fprintf(ficlog,"Age %d", iage);
4789: }
1.265 brouard 4790: for(s1=1; s1 <=nlstate ; s1++){
4791: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4792: pp[s1] += freq[s1][m][iage];
1.251 brouard 4793: }
1.265 brouard 4794: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4795: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4796: pos += freq[s1][m][iage];
4797: if(pp[s1]>=1.e-10){
1.251 brouard 4798: if(first==1){
1.265 brouard 4799: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4800: }
1.265 brouard 4801: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4802: }else{
4803: if(first==1)
1.265 brouard 4804: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4805: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4806: }
4807: }
4808:
1.265 brouard 4809: for(s1=1; s1 <=nlstate ; s1++){
4810: /* posprop[s1]=0; */
4811: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4812: pp[s1] += freq[s1][m][iage];
4813: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4814:
4815: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4816: pos += pp[s1]; /* pos is the total number of transitions until this age */
4817: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4818: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4819: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4820: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4821: }
4822:
4823: /* Writing ficresp */
4824: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4825: if( iage <= iagemax){
4826: fprintf(ficresp," %d",iage);
4827: }
4828: }else if( nj==2){
4829: if( iage <= iagemax){
4830: fprintf(ficresp," %d",iage);
4831: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4832: }
1.240 brouard 4833: }
1.265 brouard 4834: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4835: if(pos>=1.e-5){
1.251 brouard 4836: if(first==1)
1.265 brouard 4837: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4838: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4839: }else{
4840: if(first==1)
1.265 brouard 4841: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4842: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4843: }
4844: if( iage <= iagemax){
4845: if(pos>=1.e-5){
1.265 brouard 4846: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4847: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4848: }else if( nj==2){
4849: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4850: }
4851: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4852: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4853: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4854: } else{
4855: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4856: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4857: }
1.240 brouard 4858: }
1.265 brouard 4859: pospropt[s1] +=posprop[s1];
4860: } /* end loop s1 */
1.251 brouard 4861: /* pospropt=0.; */
1.265 brouard 4862: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4863: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4864: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4865: if(first==1){
1.265 brouard 4866: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4867: }
1.265 brouard 4868: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4869: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4870: }
1.265 brouard 4871: if(s1!=0 && m!=0)
4872: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4873: }
1.265 brouard 4874: } /* end loop s1 */
1.251 brouard 4875: posproptt=0.;
1.265 brouard 4876: for(s1=1; s1 <=nlstate; s1++){
4877: posproptt += pospropt[s1];
1.251 brouard 4878: }
4879: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4880: fprintf(ficresphtm,"</tr>\n");
4881: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4882: if(iage <= iagemax)
4883: fprintf(ficresp,"\n");
1.240 brouard 4884: }
1.251 brouard 4885: if(first==1)
4886: printf("Others in log...\n");
4887: fprintf(ficlog,"\n");
4888: } /* end loop age iage */
1.265 brouard 4889:
1.251 brouard 4890: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4891: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4892: if(posproptt < 1.e-5){
1.265 brouard 4893: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4894: }else{
1.265 brouard 4895: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4896: }
1.226 brouard 4897: }
1.251 brouard 4898: fprintf(ficresphtm,"</tr>\n");
4899: fprintf(ficresphtm,"</table>\n");
4900: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4901: if(posproptt < 1.e-5){
1.251 brouard 4902: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4903: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4904: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4905: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4906: invalidvarcomb[j1]=1;
1.226 brouard 4907: }else{
1.251 brouard 4908: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4909: invalidvarcomb[j1]=0;
1.226 brouard 4910: }
1.251 brouard 4911: fprintf(ficresphtmfr,"</table>\n");
4912: fprintf(ficlog,"\n");
4913: if(j!=0){
4914: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4915: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4916: for(k=1; k <=(nlstate+ndeath); k++){
4917: if (k != i) {
1.265 brouard 4918: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4919: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4920: if(j1==1){ /* All dummy covariates to zero */
4921: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4922: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4923: printf("%d%d ",i,k);
4924: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4925: 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]));
4926: 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]));
4927: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4928: }
1.253 brouard 4929: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4930: for(iage=iagemin; iage <= iagemax+3; iage++){
4931: x[iage]= (double)iage;
4932: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4933: /* 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 4934: }
1.268 brouard 4935: /* Some are not finite, but linreg will ignore these ages */
4936: no=0;
1.253 brouard 4937: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4938: pstart[s1]=b;
4939: pstart[s1-1]=a;
1.252 brouard 4940: }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 */
4941: 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]);
4942: 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 4943: 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 4944: printf("%d%d ",i,k);
4945: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4946: 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 4947: }else{ /* Other cases, like quantitative fixed or varying covariates */
4948: ;
4949: }
4950: /* printf("%12.7f )", param[i][jj][k]); */
4951: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4952: s1++;
1.251 brouard 4953: } /* end jj */
4954: } /* end k!= i */
4955: } /* end k */
1.265 brouard 4956: } /* end i, s1 */
1.251 brouard 4957: } /* end j !=0 */
4958: } /* end selected combination of covariate j1 */
4959: if(j==0){ /* We can estimate starting values from the occurences in each case */
4960: printf("#Freqsummary: Starting values for the constants:\n");
4961: fprintf(ficlog,"\n");
1.265 brouard 4962: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4963: for(k=1; k <=(nlstate+ndeath); k++){
4964: if (k != i) {
4965: printf("%d%d ",i,k);
4966: fprintf(ficlog,"%d%d ",i,k);
4967: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4968: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4969: if(jj==1){ /* Age has to be done */
1.265 brouard 4970: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4971: 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]));
4972: 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 4973: }
4974: /* printf("%12.7f )", param[i][jj][k]); */
4975: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4976: s1++;
1.250 brouard 4977: }
1.251 brouard 4978: printf("\n");
4979: fprintf(ficlog,"\n");
1.250 brouard 4980: }
4981: }
1.284 brouard 4982: } /* end of state i */
1.251 brouard 4983: printf("#Freqsummary\n");
4984: fprintf(ficlog,"\n");
1.265 brouard 4985: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4986: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4987: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4988: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4989: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4990: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
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]); */
1.251 brouard 4993: /* } */
4994: }
1.265 brouard 4995: } /* end loop s1 */
1.251 brouard 4996:
4997: printf("\n");
4998: fprintf(ficlog,"\n");
4999: } /* end j=0 */
1.249 brouard 5000: } /* end j */
1.252 brouard 5001:
1.253 brouard 5002: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5003: for(i=1, jk=1; i <=nlstate; i++){
5004: for(j=1; j <=nlstate+ndeath; j++){
5005: if(j!=i){
5006: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5007: printf("%1d%1d",i,j);
5008: fprintf(ficparo,"%1d%1d",i,j);
5009: for(k=1; k<=ncovmodel;k++){
5010: /* printf(" %lf",param[i][j][k]); */
5011: /* fprintf(ficparo," %lf",param[i][j][k]); */
5012: p[jk]=pstart[jk];
5013: printf(" %f ",pstart[jk]);
5014: fprintf(ficparo," %f ",pstart[jk]);
5015: jk++;
5016: }
5017: printf("\n");
5018: fprintf(ficparo,"\n");
5019: }
5020: }
5021: }
5022: } /* end mle=-2 */
1.226 brouard 5023: dateintmean=dateintsum/k2cpt;
1.296 brouard 5024: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5025:
1.226 brouard 5026: fclose(ficresp);
5027: fclose(ficresphtm);
5028: fclose(ficresphtmfr);
1.283 brouard 5029: free_vector(idq,1,nqfveff);
1.226 brouard 5030: free_vector(meanq,1,nqfveff);
1.284 brouard 5031: free_vector(stdq,1,nqfveff);
1.226 brouard 5032: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5033: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5034: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5035: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5036: free_vector(pospropt,1,nlstate);
5037: free_vector(posprop,1,nlstate);
1.251 brouard 5038: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5039: free_vector(pp,1,nlstate);
5040: /* End of freqsummary */
5041: }
1.126 brouard 5042:
1.268 brouard 5043: /* Simple linear regression */
5044: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5045:
5046: /* y=a+bx regression */
5047: double sumx = 0.0; /* sum of x */
5048: double sumx2 = 0.0; /* sum of x**2 */
5049: double sumxy = 0.0; /* sum of x * y */
5050: double sumy = 0.0; /* sum of y */
5051: double sumy2 = 0.0; /* sum of y**2 */
5052: double sume2 = 0.0; /* sum of square or residuals */
5053: double yhat;
5054:
5055: double denom=0;
5056: int i;
5057: int ne=*no;
5058:
5059: for ( i=ifi, ne=0;i<=ila;i++) {
5060: if(!isfinite(x[i]) || !isfinite(y[i])){
5061: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5062: continue;
5063: }
5064: ne=ne+1;
5065: sumx += x[i];
5066: sumx2 += x[i]*x[i];
5067: sumxy += x[i] * y[i];
5068: sumy += y[i];
5069: sumy2 += y[i]*y[i];
5070: denom = (ne * sumx2 - sumx*sumx);
5071: /* 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); */
5072: }
5073:
5074: denom = (ne * sumx2 - sumx*sumx);
5075: if (denom == 0) {
5076: // vertical, slope m is infinity
5077: *b = INFINITY;
5078: *a = 0;
5079: if (r) *r = 0;
5080: return 1;
5081: }
5082:
5083: *b = (ne * sumxy - sumx * sumy) / denom;
5084: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5085: if (r!=NULL) {
5086: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5087: sqrt((sumx2 - sumx*sumx/ne) *
5088: (sumy2 - sumy*sumy/ne));
5089: }
5090: *no=ne;
5091: for ( i=ifi, ne=0;i<=ila;i++) {
5092: if(!isfinite(x[i]) || !isfinite(y[i])){
5093: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5094: continue;
5095: }
5096: ne=ne+1;
5097: yhat = y[i] - *a -*b* x[i];
5098: sume2 += yhat * yhat ;
5099:
5100: denom = (ne * sumx2 - sumx*sumx);
5101: /* 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); */
5102: }
5103: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5104: *sa= *sb * sqrt(sumx2/ne);
5105:
5106: return 0;
5107: }
5108:
1.126 brouard 5109: /************ Prevalence ********************/
1.227 brouard 5110: 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)
5111: {
5112: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5113: in each health status at the date of interview (if between dateprev1 and dateprev2).
5114: We still use firstpass and lastpass as another selection.
5115: */
1.126 brouard 5116:
1.227 brouard 5117: int i, m, jk, j1, bool, z1,j, iv;
5118: int mi; /* Effective wave */
5119: int iage;
5120: double agebegin, ageend;
5121:
5122: double **prop;
5123: double posprop;
5124: double y2; /* in fractional years */
5125: int iagemin, iagemax;
5126: int first; /** to stop verbosity which is redirected to log file */
5127:
5128: iagemin= (int) agemin;
5129: iagemax= (int) agemax;
5130: /*pp=vector(1,nlstate);*/
1.251 brouard 5131: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5132: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5133: j1=0;
1.222 brouard 5134:
1.227 brouard 5135: /*j=cptcoveff;*/
5136: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5137:
1.288 brouard 5138: first=0;
1.227 brouard 5139: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5140: for (i=1; i<=nlstate; i++)
1.251 brouard 5141: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5142: prop[i][iage]=0.0;
5143: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5144: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5145: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5146:
5147: for (i=1; i<=imx; i++) { /* Each individual */
5148: bool=1;
5149: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5150: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5151: m=mw[mi][i];
5152: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5153: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5154: for (z1=1; z1<=cptcoveff; z1++){
5155: if( Fixed[Tmodelind[z1]]==1){
5156: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5157: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5158: bool=0;
5159: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5160: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5161: bool=0;
5162: }
5163: }
5164: if(bool==1){ /* Otherwise we skip that wave/person */
5165: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5166: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5167: if(m >=firstpass && m <=lastpass){
5168: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5169: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5170: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5171: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5172: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5173: 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);
5174: exit(1);
5175: }
5176: if (s[m][i]>0 && s[m][i]<=nlstate) {
5177: /*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]]);*/
5178: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5179: prop[s[m][i]][iagemax+3] += weight[i];
5180: } /* end valid statuses */
5181: } /* end selection of dates */
5182: } /* end selection of waves */
5183: } /* end bool */
5184: } /* end wave */
5185: } /* end individual */
5186: for(i=iagemin; i <= iagemax+3; i++){
5187: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5188: posprop += prop[jk][i];
5189: }
5190:
5191: for(jk=1; jk <=nlstate ; jk++){
5192: if( i <= iagemax){
5193: if(posprop>=1.e-5){
5194: probs[i][jk][j1]= prop[jk][i]/posprop;
5195: } else{
1.288 brouard 5196: if(!first){
5197: first=1;
1.266 brouard 5198: 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]);
5199: }else{
1.288 brouard 5200: 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 5201: }
5202: }
5203: }
5204: }/* end jk */
5205: }/* end i */
1.222 brouard 5206: /*} *//* end i1 */
1.227 brouard 5207: } /* end j1 */
1.222 brouard 5208:
1.227 brouard 5209: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5210: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5211: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5212: } /* End of prevalence */
1.126 brouard 5213:
5214: /************* Waves Concatenation ***************/
5215:
5216: 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)
5217: {
1.298 brouard 5218: /* 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 5219: Death is a valid wave (if date is known).
5220: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5221: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5222: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5223: */
1.126 brouard 5224:
1.224 brouard 5225: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5226: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5227: double sum=0., jmean=0.;*/
1.224 brouard 5228: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5229: int j, k=0,jk, ju, jl;
5230: double sum=0.;
5231: first=0;
1.214 brouard 5232: firstwo=0;
1.217 brouard 5233: firsthree=0;
1.218 brouard 5234: firstfour=0;
1.164 brouard 5235: jmin=100000;
1.126 brouard 5236: jmax=-1;
5237: jmean=0.;
1.224 brouard 5238:
5239: /* Treating live states */
1.214 brouard 5240: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5241: mi=0; /* First valid wave */
1.227 brouard 5242: mli=0; /* Last valid wave */
1.126 brouard 5243: m=firstpass;
1.214 brouard 5244: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5245: 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 */
5246: mli=m-1;/* mw[++mi][i]=m-1; */
5247: }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 */
5248: mw[++mi][i]=m;
5249: mli=m;
1.224 brouard 5250: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5251: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5252: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5253: }
1.227 brouard 5254: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5255: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5256: break;
1.224 brouard 5257: #else
1.227 brouard 5258: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5259: if(firsthree == 0){
1.262 brouard 5260: 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 5261: firsthree=1;
5262: }
1.262 brouard 5263: 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 5264: mw[++mi][i]=m;
5265: mli=m;
5266: }
5267: if(s[m][i]==-2){ /* Vital status is really unknown */
5268: nbwarn++;
5269: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5270: 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);
5271: 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);
5272: }
5273: break;
5274: }
5275: break;
1.224 brouard 5276: #endif
1.227 brouard 5277: }/* End m >= lastpass */
1.126 brouard 5278: }/* end while */
1.224 brouard 5279:
1.227 brouard 5280: /* 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 5281: /* After last pass */
1.224 brouard 5282: /* Treating death states */
1.214 brouard 5283: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5284: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5285: /* } */
1.126 brouard 5286: mi++; /* Death is another wave */
5287: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5288: /* Only death is a correct wave */
1.126 brouard 5289: mw[mi][i]=m;
1.257 brouard 5290: } /* else not in a death state */
1.224 brouard 5291: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5292: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5293: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5294: 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 */
5295: nbwarn++;
5296: if(firstfiv==0){
5297: 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 );
5298: firstfiv=1;
5299: }else{
5300: 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 );
5301: }
5302: }else{ /* Death occured afer last wave potential bias */
5303: nberr++;
5304: if(firstwo==0){
1.257 brouard 5305: 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 5306: firstwo=1;
5307: }
1.257 brouard 5308: 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 5309: }
1.257 brouard 5310: }else{ /* if date of interview is unknown */
1.227 brouard 5311: /* death is known but not confirmed by death status at any wave */
5312: if(firstfour==0){
5313: 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 );
5314: firstfour=1;
5315: }
5316: 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 5317: }
1.224 brouard 5318: } /* end if date of death is known */
5319: #endif
5320: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5321: /* wav[i]=mw[mi][i]; */
1.126 brouard 5322: if(mi==0){
5323: nbwarn++;
5324: if(first==0){
1.227 brouard 5325: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5326: first=1;
1.126 brouard 5327: }
5328: if(first==1){
1.227 brouard 5329: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5330: }
5331: } /* end mi==0 */
5332: } /* End individuals */
1.214 brouard 5333: /* wav and mw are no more changed */
1.223 brouard 5334:
1.214 brouard 5335:
1.126 brouard 5336: for(i=1; i<=imx; i++){
5337: for(mi=1; mi<wav[i];mi++){
5338: if (stepm <=0)
1.227 brouard 5339: dh[mi][i]=1;
1.126 brouard 5340: else{
1.260 brouard 5341: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5342: if (agedc[i] < 2*AGESUP) {
5343: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5344: if(j==0) j=1; /* Survives at least one month after exam */
5345: else if(j<0){
5346: nberr++;
5347: 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]);
5348: j=1; /* Temporary Dangerous patch */
5349: 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);
5350: 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]);
5351: 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);
5352: }
5353: k=k+1;
5354: if (j >= jmax){
5355: jmax=j;
5356: ijmax=i;
5357: }
5358: if (j <= jmin){
5359: jmin=j;
5360: ijmin=i;
5361: }
5362: sum=sum+j;
5363: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5364: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5365: }
5366: }
5367: else{
5368: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5369: /* 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 5370:
1.227 brouard 5371: k=k+1;
5372: if (j >= jmax) {
5373: jmax=j;
5374: ijmax=i;
5375: }
5376: else if (j <= jmin){
5377: jmin=j;
5378: ijmin=i;
5379: }
5380: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5381: /*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]);*/
5382: if(j<0){
5383: nberr++;
5384: 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]);
5385: 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]);
5386: }
5387: sum=sum+j;
5388: }
5389: jk= j/stepm;
5390: jl= j -jk*stepm;
5391: ju= j -(jk+1)*stepm;
5392: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5393: if(jl==0){
5394: dh[mi][i]=jk;
5395: bh[mi][i]=0;
5396: }else{ /* We want a negative bias in order to only have interpolation ie
5397: * to avoid the price of an extra matrix product in likelihood */
5398: dh[mi][i]=jk+1;
5399: bh[mi][i]=ju;
5400: }
5401: }else{
5402: if(jl <= -ju){
5403: dh[mi][i]=jk;
5404: bh[mi][i]=jl; /* bias is positive if real duration
5405: * is higher than the multiple of stepm and negative otherwise.
5406: */
5407: }
5408: else{
5409: dh[mi][i]=jk+1;
5410: bh[mi][i]=ju;
5411: }
5412: if(dh[mi][i]==0){
5413: dh[mi][i]=1; /* At least one step */
5414: bh[mi][i]=ju; /* At least one step */
5415: /* 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);*/
5416: }
5417: } /* end if mle */
1.126 brouard 5418: }
5419: } /* end wave */
5420: }
5421: jmean=sum/k;
5422: 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 5423: 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 5424: }
1.126 brouard 5425:
5426: /*********** Tricode ****************************/
1.220 brouard 5427: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5428: {
5429: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5430: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5431: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5432: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5433: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5434: */
1.130 brouard 5435:
1.242 brouard 5436: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5437: int modmaxcovj=0; /* Modality max of covariates j */
5438: int cptcode=0; /* Modality max of covariates j */
5439: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5440:
5441:
1.242 brouard 5442: /* cptcoveff=0; */
5443: /* *cptcov=0; */
1.126 brouard 5444:
1.242 brouard 5445: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5446: for (k=1; k <= maxncov; k++)
5447: for(j=1; j<=2; j++)
5448: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5449:
1.242 brouard 5450: /* Loop on covariates without age and products and no quantitative variable */
5451: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5452: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5453: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5454: switch(Fixed[k]) {
5455: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5456: 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*/
5457: ij=(int)(covar[Tvar[k]][i]);
5458: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5459: * If product of Vn*Vm, still boolean *:
5460: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5461: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5462: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5463: modality of the nth covariate of individual i. */
5464: if (ij > modmaxcovj)
5465: modmaxcovj=ij;
5466: else if (ij < modmincovj)
5467: modmincovj=ij;
1.287 brouard 5468: if (ij <0 || ij >1 ){
5469: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5470: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5471: }
5472: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5473: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5474: exit(1);
5475: }else
5476: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5477: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5478: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5479: /* getting the maximum value of the modality of the covariate
5480: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5481: female ies 1, then modmaxcovj=1.
5482: */
5483: } /* end for loop on individuals i */
5484: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5485: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5486: cptcode=modmaxcovj;
5487: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5488: /*for (i=0; i<=cptcode; i++) {*/
5489: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5490: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5491: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5492: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5493: if( j != -1){
5494: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5495: covariate for which somebody answered excluding
5496: undefined. Usually 2: 0 and 1. */
5497: }
5498: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5499: covariate for which somebody answered including
5500: undefined. Usually 3: -1, 0 and 1. */
5501: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5502: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5503: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5504:
1.242 brouard 5505: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5506: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5507: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5508: /* modmincovj=3; modmaxcovj = 7; */
5509: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5510: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5511: /* defining two dummy variables: variables V1_1 and V1_2.*/
5512: /* nbcode[Tvar[j]][ij]=k; */
5513: /* nbcode[Tvar[j]][1]=0; */
5514: /* nbcode[Tvar[j]][2]=1; */
5515: /* nbcode[Tvar[j]][3]=2; */
5516: /* To be continued (not working yet). */
5517: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5518:
5519: /* 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*/
5520: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5521: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5522: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5523: /*, could be restored in the future */
5524: 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 5525: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5526: break;
5527: }
5528: ij++;
1.287 brouard 5529: 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 5530: cptcode = ij; /* New max modality for covar j */
5531: } /* end of loop on modality i=-1 to 1 or more */
5532: break;
5533: case 1: /* Testing on varying covariate, could be simple and
5534: * should look at waves or product of fixed *
5535: * varying. No time to test -1, assuming 0 and 1 only */
5536: ij=0;
5537: for(i=0; i<=1;i++){
5538: nbcode[Tvar[k]][++ij]=i;
5539: }
5540: break;
5541: default:
5542: break;
5543: } /* end switch */
5544: } /* end dummy test */
1.287 brouard 5545: } /* 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 5546:
5547: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5548: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5549: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5550: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5551: 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 */
5552: 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 */
5553: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5554: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5555:
5556: ij=0;
5557: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5558: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5559: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5560: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5561: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5562: /* If product not in single variable we don't print results */
5563: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5564: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5565: 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*/
5566: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5567: 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 */
5568: if(Fixed[k]!=0)
5569: anyvaryingduminmodel=1;
5570: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5571: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5572: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5573: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5574: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5575: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5576: }
5577: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5578: /* ij--; */
5579: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5580: *cptcov=ij; /*Number of total real effective covariates: effective
5581: * because they can be excluded from the model and real
5582: * if in the model but excluded because missing values, but how to get k from ij?*/
5583: for(j=ij+1; j<= cptcovt; j++){
5584: Tvaraff[j]=0;
5585: Tmodelind[j]=0;
5586: }
5587: for(j=ntveff+1; j<= cptcovt; j++){
5588: TmodelInvind[j]=0;
5589: }
5590: /* To be sorted */
5591: ;
5592: }
1.126 brouard 5593:
1.145 brouard 5594:
1.126 brouard 5595: /*********** Health Expectancies ****************/
5596:
1.235 brouard 5597: 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 5598:
5599: {
5600: /* Health expectancies, no variances */
1.164 brouard 5601: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5602: int nhstepma, nstepma; /* Decreasing with age */
5603: double age, agelim, hf;
5604: double ***p3mat;
5605: double eip;
5606:
1.238 brouard 5607: /* pstamp(ficreseij); */
1.126 brouard 5608: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5609: fprintf(ficreseij,"# Age");
5610: for(i=1; i<=nlstate;i++){
5611: for(j=1; j<=nlstate;j++){
5612: fprintf(ficreseij," e%1d%1d ",i,j);
5613: }
5614: fprintf(ficreseij," e%1d. ",i);
5615: }
5616: fprintf(ficreseij,"\n");
5617:
5618:
5619: if(estepm < stepm){
5620: printf ("Problem %d lower than %d\n",estepm, stepm);
5621: }
5622: else hstepm=estepm;
5623: /* We compute the life expectancy from trapezoids spaced every estepm months
5624: * This is mainly to measure the difference between two models: for example
5625: * if stepm=24 months pijx are given only every 2 years and by summing them
5626: * we are calculating an estimate of the Life Expectancy assuming a linear
5627: * progression in between and thus overestimating or underestimating according
5628: * to the curvature of the survival function. If, for the same date, we
5629: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5630: * to compare the new estimate of Life expectancy with the same linear
5631: * hypothesis. A more precise result, taking into account a more precise
5632: * curvature will be obtained if estepm is as small as stepm. */
5633:
5634: /* For example we decided to compute the life expectancy with the smallest unit */
5635: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5636: nhstepm is the number of hstepm from age to agelim
5637: nstepm is the number of stepm from age to agelin.
1.270 brouard 5638: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5639: and note for a fixed period like estepm months */
5640: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5641: survival function given by stepm (the optimization length). Unfortunately it
5642: means that if the survival funtion is printed only each two years of age and if
5643: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5644: results. So we changed our mind and took the option of the best precision.
5645: */
5646: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5647:
5648: agelim=AGESUP;
5649: /* If stepm=6 months */
5650: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5651: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5652:
5653: /* nhstepm age range expressed in number of stepm */
5654: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5655: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5656: /* if (stepm >= YEARM) hstepm=1;*/
5657: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5658: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5659:
5660: for (age=bage; age<=fage; age ++){
5661: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5662: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5663: /* if (stepm >= YEARM) hstepm=1;*/
5664: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5665:
5666: /* If stepm=6 months */
5667: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5668: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5669:
1.235 brouard 5670: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5671:
5672: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5673:
5674: printf("%d|",(int)age);fflush(stdout);
5675: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5676:
5677: /* Computing expectancies */
5678: for(i=1; i<=nlstate;i++)
5679: for(j=1; j<=nlstate;j++)
5680: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5681: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5682:
5683: /* 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]);*/
5684:
5685: }
5686:
5687: fprintf(ficreseij,"%3.0f",age );
5688: for(i=1; i<=nlstate;i++){
5689: eip=0;
5690: for(j=1; j<=nlstate;j++){
5691: eip +=eij[i][j][(int)age];
5692: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5693: }
5694: fprintf(ficreseij,"%9.4f", eip );
5695: }
5696: fprintf(ficreseij,"\n");
5697:
5698: }
5699: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5700: printf("\n");
5701: fprintf(ficlog,"\n");
5702:
5703: }
5704:
1.235 brouard 5705: 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 5706:
5707: {
5708: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5709: to initial status i, ei. .
1.126 brouard 5710: */
5711: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5712: int nhstepma, nstepma; /* Decreasing with age */
5713: double age, agelim, hf;
5714: double ***p3matp, ***p3matm, ***varhe;
5715: double **dnewm,**doldm;
5716: double *xp, *xm;
5717: double **gp, **gm;
5718: double ***gradg, ***trgradg;
5719: int theta;
5720:
5721: double eip, vip;
5722:
5723: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5724: xp=vector(1,npar);
5725: xm=vector(1,npar);
5726: dnewm=matrix(1,nlstate*nlstate,1,npar);
5727: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5728:
5729: pstamp(ficresstdeij);
5730: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5731: fprintf(ficresstdeij,"# Age");
5732: for(i=1; i<=nlstate;i++){
5733: for(j=1; j<=nlstate;j++)
5734: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5735: fprintf(ficresstdeij," e%1d. ",i);
5736: }
5737: fprintf(ficresstdeij,"\n");
5738:
5739: pstamp(ficrescveij);
5740: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5741: fprintf(ficrescveij,"# Age");
5742: for(i=1; i<=nlstate;i++)
5743: for(j=1; j<=nlstate;j++){
5744: cptj= (j-1)*nlstate+i;
5745: for(i2=1; i2<=nlstate;i2++)
5746: for(j2=1; j2<=nlstate;j2++){
5747: cptj2= (j2-1)*nlstate+i2;
5748: if(cptj2 <= cptj)
5749: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5750: }
5751: }
5752: fprintf(ficrescveij,"\n");
5753:
5754: if(estepm < stepm){
5755: printf ("Problem %d lower than %d\n",estepm, stepm);
5756: }
5757: else hstepm=estepm;
5758: /* We compute the life expectancy from trapezoids spaced every estepm months
5759: * This is mainly to measure the difference between two models: for example
5760: * if stepm=24 months pijx are given only every 2 years and by summing them
5761: * we are calculating an estimate of the Life Expectancy assuming a linear
5762: * progression in between and thus overestimating or underestimating according
5763: * to the curvature of the survival function. If, for the same date, we
5764: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5765: * to compare the new estimate of Life expectancy with the same linear
5766: * hypothesis. A more precise result, taking into account a more precise
5767: * curvature will be obtained if estepm is as small as stepm. */
5768:
5769: /* For example we decided to compute the life expectancy with the smallest unit */
5770: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5771: nhstepm is the number of hstepm from age to agelim
5772: nstepm is the number of stepm from age to agelin.
5773: Look at hpijx to understand the reason of that which relies in memory size
5774: and note for a fixed period like estepm months */
5775: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5776: survival function given by stepm (the optimization length). Unfortunately it
5777: means that if the survival funtion is printed only each two years of age and if
5778: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5779: results. So we changed our mind and took the option of the best precision.
5780: */
5781: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5782:
5783: /* If stepm=6 months */
5784: /* nhstepm age range expressed in number of stepm */
5785: agelim=AGESUP;
5786: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5787: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5788: /* if (stepm >= YEARM) hstepm=1;*/
5789: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5790:
5791: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5792: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5793: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5794: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5795: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5796: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5797:
5798: for (age=bage; age<=fage; age ++){
5799: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5800: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5801: /* if (stepm >= YEARM) hstepm=1;*/
5802: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5803:
1.126 brouard 5804: /* If stepm=6 months */
5805: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5806: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5807:
5808: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5809:
1.126 brouard 5810: /* Computing Variances of health expectancies */
5811: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5812: decrease memory allocation */
5813: for(theta=1; theta <=npar; theta++){
5814: for(i=1; i<=npar; i++){
1.222 brouard 5815: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5816: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5817: }
1.235 brouard 5818: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5819: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5820:
1.126 brouard 5821: for(j=1; j<= nlstate; j++){
1.222 brouard 5822: for(i=1; i<=nlstate; i++){
5823: for(h=0; h<=nhstepm-1; h++){
5824: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5825: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5826: }
5827: }
1.126 brouard 5828: }
1.218 brouard 5829:
1.126 brouard 5830: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5831: for(h=0; h<=nhstepm-1; h++){
5832: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5833: }
1.126 brouard 5834: }/* End theta */
5835:
5836:
5837: for(h=0; h<=nhstepm-1; h++)
5838: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5839: for(theta=1; theta <=npar; theta++)
5840: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5841:
1.218 brouard 5842:
1.222 brouard 5843: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5844: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5845: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5846:
1.222 brouard 5847: printf("%d|",(int)age);fflush(stdout);
5848: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5849: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5850: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5851: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5852: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5853: for(ij=1;ij<=nlstate*nlstate;ij++)
5854: for(ji=1;ji<=nlstate*nlstate;ji++)
5855: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5856: }
5857: }
1.218 brouard 5858:
1.126 brouard 5859: /* Computing expectancies */
1.235 brouard 5860: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5861: for(i=1; i<=nlstate;i++)
5862: for(j=1; j<=nlstate;j++)
1.222 brouard 5863: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5864: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5865:
1.222 brouard 5866: /* 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 5867:
1.222 brouard 5868: }
1.269 brouard 5869:
5870: /* Standard deviation of expectancies ij */
1.126 brouard 5871: fprintf(ficresstdeij,"%3.0f",age );
5872: for(i=1; i<=nlstate;i++){
5873: eip=0.;
5874: vip=0.;
5875: for(j=1; j<=nlstate;j++){
1.222 brouard 5876: eip += eij[i][j][(int)age];
5877: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5878: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5879: 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 5880: }
5881: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5882: }
5883: fprintf(ficresstdeij,"\n");
1.218 brouard 5884:
1.269 brouard 5885: /* Variance of expectancies ij */
1.126 brouard 5886: fprintf(ficrescveij,"%3.0f",age );
5887: for(i=1; i<=nlstate;i++)
5888: for(j=1; j<=nlstate;j++){
1.222 brouard 5889: cptj= (j-1)*nlstate+i;
5890: for(i2=1; i2<=nlstate;i2++)
5891: for(j2=1; j2<=nlstate;j2++){
5892: cptj2= (j2-1)*nlstate+i2;
5893: if(cptj2 <= cptj)
5894: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5895: }
1.126 brouard 5896: }
5897: fprintf(ficrescveij,"\n");
1.218 brouard 5898:
1.126 brouard 5899: }
5900: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5901: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5902: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5903: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5904: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5905: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5906: printf("\n");
5907: fprintf(ficlog,"\n");
1.218 brouard 5908:
1.126 brouard 5909: free_vector(xm,1,npar);
5910: free_vector(xp,1,npar);
5911: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5912: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5913: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5914: }
1.218 brouard 5915:
1.126 brouard 5916: /************ Variance ******************/
1.235 brouard 5917: 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 5918: {
1.279 brouard 5919: /** Variance of health expectancies
5920: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5921: * double **newm;
5922: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5923: */
1.218 brouard 5924:
5925: /* int movingaverage(); */
5926: double **dnewm,**doldm;
5927: double **dnewmp,**doldmp;
5928: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5929: int first=0;
1.218 brouard 5930: int k;
5931: double *xp;
1.279 brouard 5932: double **gp, **gm; /**< for var eij */
5933: double ***gradg, ***trgradg; /**< for var eij */
5934: double **gradgp, **trgradgp; /**< for var p point j */
5935: double *gpp, *gmp; /**< for var p point j */
5936: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5937: double ***p3mat;
5938: double age,agelim, hf;
5939: /* double ***mobaverage; */
5940: int theta;
5941: char digit[4];
5942: char digitp[25];
5943:
5944: char fileresprobmorprev[FILENAMELENGTH];
5945:
5946: if(popbased==1){
5947: if(mobilav!=0)
5948: strcpy(digitp,"-POPULBASED-MOBILAV_");
5949: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5950: }
5951: else
5952: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5953:
1.218 brouard 5954: /* if (mobilav!=0) { */
5955: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5956: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5957: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5958: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5959: /* } */
5960: /* } */
5961:
5962: strcpy(fileresprobmorprev,"PRMORPREV-");
5963: sprintf(digit,"%-d",ij);
5964: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5965: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5966: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5967: strcat(fileresprobmorprev,fileresu);
5968: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5969: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5970: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5971: }
5972: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5973: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5974: pstamp(ficresprobmorprev);
5975: 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 5976: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5977: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5978: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5979: }
5980: for(j=1;j<=cptcoveff;j++)
5981: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5982: fprintf(ficresprobmorprev,"\n");
5983:
1.218 brouard 5984: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5985: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5986: fprintf(ficresprobmorprev," p.%-d SE",j);
5987: for(i=1; i<=nlstate;i++)
5988: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5989: }
5990: fprintf(ficresprobmorprev,"\n");
5991:
5992: fprintf(ficgp,"\n# Routine varevsij");
5993: fprintf(ficgp,"\nunset title \n");
5994: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5995: 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");
5996: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5997:
1.218 brouard 5998: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5999: pstamp(ficresvij);
6000: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6001: if(popbased==1)
6002: 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);
6003: else
6004: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6005: fprintf(ficresvij,"# Age");
6006: for(i=1; i<=nlstate;i++)
6007: for(j=1; j<=nlstate;j++)
6008: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6009: fprintf(ficresvij,"\n");
6010:
6011: xp=vector(1,npar);
6012: dnewm=matrix(1,nlstate,1,npar);
6013: doldm=matrix(1,nlstate,1,nlstate);
6014: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6015: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6016:
6017: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6018: gpp=vector(nlstate+1,nlstate+ndeath);
6019: gmp=vector(nlstate+1,nlstate+ndeath);
6020: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6021:
1.218 brouard 6022: if(estepm < stepm){
6023: printf ("Problem %d lower than %d\n",estepm, stepm);
6024: }
6025: else hstepm=estepm;
6026: /* For example we decided to compute the life expectancy with the smallest unit */
6027: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6028: nhstepm is the number of hstepm from age to agelim
6029: nstepm is the number of stepm from age to agelim.
6030: Look at function hpijx to understand why because of memory size limitations,
6031: we decided (b) to get a life expectancy respecting the most precise curvature of the
6032: survival function given by stepm (the optimization length). Unfortunately it
6033: means that if the survival funtion is printed every two years of age and if
6034: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6035: results. So we changed our mind and took the option of the best precision.
6036: */
6037: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6038: agelim = AGESUP;
6039: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6040: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6041: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6042: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6043: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6044: gp=matrix(0,nhstepm,1,nlstate);
6045: gm=matrix(0,nhstepm,1,nlstate);
6046:
6047:
6048: for(theta=1; theta <=npar; theta++){
6049: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6050: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6051: }
1.279 brouard 6052: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6053: * returns into prlim .
1.288 brouard 6054: */
1.242 brouard 6055: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6056:
6057: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6058: if (popbased==1) {
6059: if(mobilav ==0){
6060: for(i=1; i<=nlstate;i++)
6061: prlim[i][i]=probs[(int)age][i][ij];
6062: }else{ /* mobilav */
6063: for(i=1; i<=nlstate;i++)
6064: prlim[i][i]=mobaverage[(int)age][i][ij];
6065: }
6066: }
1.295 brouard 6067: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6068: */
6069: 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 6070: /**< 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 6071: * at horizon h in state j including mortality.
6072: */
1.218 brouard 6073: for(j=1; j<= nlstate; j++){
6074: for(h=0; h<=nhstepm; h++){
6075: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6076: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6077: }
6078: }
1.279 brouard 6079: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6080: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6081: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6082: */
6083: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6084: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6085: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6086: }
6087:
6088: /* Again with minus shift */
1.218 brouard 6089:
6090: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6091: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6092:
1.242 brouard 6093: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6094:
6095: if (popbased==1) {
6096: if(mobilav ==0){
6097: for(i=1; i<=nlstate;i++)
6098: prlim[i][i]=probs[(int)age][i][ij];
6099: }else{ /* mobilav */
6100: for(i=1; i<=nlstate;i++)
6101: prlim[i][i]=mobaverage[(int)age][i][ij];
6102: }
6103: }
6104:
1.235 brouard 6105: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6106:
6107: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6108: for(h=0; h<=nhstepm; h++){
6109: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6110: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6111: }
6112: }
6113: /* This for computing probability of death (h=1 means
6114: computed over hstepm matrices product = hstepm*stepm months)
6115: as a weighted average of prlim.
6116: */
6117: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6118: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6119: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6120: }
1.279 brouard 6121: /* end shifting computations */
6122:
6123: /**< Computing gradient matrix at horizon h
6124: */
1.218 brouard 6125: for(j=1; j<= nlstate; j++) /* vareij */
6126: for(h=0; h<=nhstepm; h++){
6127: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6128: }
1.279 brouard 6129: /**< Gradient of overall mortality p.3 (or p.j)
6130: */
6131: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6132: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6133: }
6134:
6135: } /* End theta */
1.279 brouard 6136:
6137: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6138: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6139:
6140: for(h=0; h<=nhstepm; h++) /* veij */
6141: for(j=1; j<=nlstate;j++)
6142: for(theta=1; theta <=npar; theta++)
6143: trgradg[h][j][theta]=gradg[h][theta][j];
6144:
6145: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6146: for(theta=1; theta <=npar; theta++)
6147: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6148: /**< as well as its transposed matrix
6149: */
1.218 brouard 6150:
6151: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6152: for(i=1;i<=nlstate;i++)
6153: for(j=1;j<=nlstate;j++)
6154: vareij[i][j][(int)age] =0.;
1.279 brouard 6155:
6156: /* Computing trgradg by matcov by gradg at age and summing over h
6157: * and k (nhstepm) formula 15 of article
6158: * Lievre-Brouard-Heathcote
6159: */
6160:
1.218 brouard 6161: for(h=0;h<=nhstepm;h++){
6162: for(k=0;k<=nhstepm;k++){
6163: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6164: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6165: for(i=1;i<=nlstate;i++)
6166: for(j=1;j<=nlstate;j++)
6167: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6168: }
6169: }
6170:
1.279 brouard 6171: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6172: * p.j overall mortality formula 49 but computed directly because
6173: * we compute the grad (wix pijx) instead of grad (pijx),even if
6174: * wix is independent of theta.
6175: */
1.218 brouard 6176: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6177: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6178: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6179: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6180: varppt[j][i]=doldmp[j][i];
6181: /* end ppptj */
6182: /* x centered again */
6183:
1.242 brouard 6184: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6185:
6186: if (popbased==1) {
6187: if(mobilav ==0){
6188: for(i=1; i<=nlstate;i++)
6189: prlim[i][i]=probs[(int)age][i][ij];
6190: }else{ /* mobilav */
6191: for(i=1; i<=nlstate;i++)
6192: prlim[i][i]=mobaverage[(int)age][i][ij];
6193: }
6194: }
6195:
6196: /* This for computing probability of death (h=1 means
6197: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6198: as a weighted average of prlim.
6199: */
1.235 brouard 6200: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6201: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6202: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6203: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6204: }
6205: /* end probability of death */
6206:
6207: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6208: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6209: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6210: for(i=1; i<=nlstate;i++){
6211: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6212: }
6213: }
6214: fprintf(ficresprobmorprev,"\n");
6215:
6216: fprintf(ficresvij,"%.0f ",age );
6217: for(i=1; i<=nlstate;i++)
6218: for(j=1; j<=nlstate;j++){
6219: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6220: }
6221: fprintf(ficresvij,"\n");
6222: free_matrix(gp,0,nhstepm,1,nlstate);
6223: free_matrix(gm,0,nhstepm,1,nlstate);
6224: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6225: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6226: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6227: } /* End age */
6228: free_vector(gpp,nlstate+1,nlstate+ndeath);
6229: free_vector(gmp,nlstate+1,nlstate+ndeath);
6230: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6231: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6232: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6233: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6234: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6235: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6236: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6237: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6238: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6239: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6240: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6241: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6242: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6243: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6244: 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);
6245: /* 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 6246: */
1.218 brouard 6247: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6248: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6249:
1.218 brouard 6250: free_vector(xp,1,npar);
6251: free_matrix(doldm,1,nlstate,1,nlstate);
6252: free_matrix(dnewm,1,nlstate,1,npar);
6253: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6254: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6255: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6256: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6257: fclose(ficresprobmorprev);
6258: fflush(ficgp);
6259: fflush(fichtm);
6260: } /* end varevsij */
1.126 brouard 6261:
6262: /************ Variance of prevlim ******************/
1.269 brouard 6263: 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 6264: {
1.205 brouard 6265: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6266: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6267:
1.268 brouard 6268: double **dnewmpar,**doldm;
1.126 brouard 6269: int i, j, nhstepm, hstepm;
6270: double *xp;
6271: double *gp, *gm;
6272: double **gradg, **trgradg;
1.208 brouard 6273: double **mgm, **mgp;
1.126 brouard 6274: double age,agelim;
6275: int theta;
6276:
6277: pstamp(ficresvpl);
1.288 brouard 6278: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6279: fprintf(ficresvpl,"# Age ");
6280: if(nresult >=1)
6281: fprintf(ficresvpl," Result# ");
1.126 brouard 6282: for(i=1; i<=nlstate;i++)
6283: fprintf(ficresvpl," %1d-%1d",i,i);
6284: fprintf(ficresvpl,"\n");
6285:
6286: xp=vector(1,npar);
1.268 brouard 6287: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6288: doldm=matrix(1,nlstate,1,nlstate);
6289:
6290: hstepm=1*YEARM; /* Every year of age */
6291: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6292: agelim = AGESUP;
6293: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6294: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6295: if (stepm >= YEARM) hstepm=1;
6296: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6297: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6298: mgp=matrix(1,npar,1,nlstate);
6299: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6300: gp=vector(1,nlstate);
6301: gm=vector(1,nlstate);
6302:
6303: for(theta=1; theta <=npar; theta++){
6304: for(i=1; i<=npar; i++){ /* Computes gradient */
6305: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6306: }
1.288 brouard 6307: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6308: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6309: /* else */
6310: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6311: for(i=1;i<=nlstate;i++){
1.126 brouard 6312: gp[i] = prlim[i][i];
1.208 brouard 6313: mgp[theta][i] = prlim[i][i];
6314: }
1.126 brouard 6315: for(i=1; i<=npar; i++) /* Computes gradient */
6316: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6317: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6318: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6319: /* else */
6320: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6321: for(i=1;i<=nlstate;i++){
1.126 brouard 6322: gm[i] = prlim[i][i];
1.208 brouard 6323: mgm[theta][i] = prlim[i][i];
6324: }
1.126 brouard 6325: for(i=1;i<=nlstate;i++)
6326: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6327: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6328: } /* End theta */
6329:
6330: trgradg =matrix(1,nlstate,1,npar);
6331:
6332: for(j=1; j<=nlstate;j++)
6333: for(theta=1; theta <=npar; theta++)
6334: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6335: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6336: /* printf("\nmgm mgp %d ",(int)age); */
6337: /* for(j=1; j<=nlstate;j++){ */
6338: /* printf(" %d ",j); */
6339: /* for(theta=1; theta <=npar; theta++) */
6340: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6341: /* printf("\n "); */
6342: /* } */
6343: /* } */
6344: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6345: /* printf("\n gradg %d ",(int)age); */
6346: /* for(j=1; j<=nlstate;j++){ */
6347: /* printf("%d ",j); */
6348: /* for(theta=1; theta <=npar; theta++) */
6349: /* printf("%d %lf ",theta,gradg[theta][j]); */
6350: /* printf("\n "); */
6351: /* } */
6352: /* } */
1.126 brouard 6353:
6354: for(i=1;i<=nlstate;i++)
6355: varpl[i][(int)age] =0.;
1.209 brouard 6356: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6357: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6358: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6359: }else{
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: }
1.126 brouard 6363: for(i=1;i<=nlstate;i++)
6364: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6365:
6366: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6367: if(nresult >=1)
6368: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6369: for(i=1; i<=nlstate;i++){
1.126 brouard 6370: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6371: /* for(j=1;j<=nlstate;j++) */
6372: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6373: }
1.126 brouard 6374: fprintf(ficresvpl,"\n");
6375: free_vector(gp,1,nlstate);
6376: free_vector(gm,1,nlstate);
1.208 brouard 6377: free_matrix(mgm,1,npar,1,nlstate);
6378: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6379: free_matrix(gradg,1,npar,1,nlstate);
6380: free_matrix(trgradg,1,nlstate,1,npar);
6381: } /* End age */
6382:
6383: free_vector(xp,1,npar);
6384: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6385: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6386:
6387: }
6388:
6389:
6390: /************ Variance of backprevalence limit ******************/
1.269 brouard 6391: 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 6392: {
6393: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6394: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6395:
6396: double **dnewmpar,**doldm;
6397: int i, j, nhstepm, hstepm;
6398: double *xp;
6399: double *gp, *gm;
6400: double **gradg, **trgradg;
6401: double **mgm, **mgp;
6402: double age,agelim;
6403: int theta;
6404:
6405: pstamp(ficresvbl);
6406: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6407: fprintf(ficresvbl,"# Age ");
6408: if(nresult >=1)
6409: fprintf(ficresvbl," Result# ");
6410: for(i=1; i<=nlstate;i++)
6411: fprintf(ficresvbl," %1d-%1d",i,i);
6412: fprintf(ficresvbl,"\n");
6413:
6414: xp=vector(1,npar);
6415: dnewmpar=matrix(1,nlstate,1,npar);
6416: doldm=matrix(1,nlstate,1,nlstate);
6417:
6418: hstepm=1*YEARM; /* Every year of age */
6419: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6420: agelim = AGEINF;
6421: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6422: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6423: if (stepm >= YEARM) hstepm=1;
6424: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6425: gradg=matrix(1,npar,1,nlstate);
6426: mgp=matrix(1,npar,1,nlstate);
6427: mgm=matrix(1,npar,1,nlstate);
6428: gp=vector(1,nlstate);
6429: gm=vector(1,nlstate);
6430:
6431: for(theta=1; theta <=npar; theta++){
6432: for(i=1; i<=npar; i++){ /* Computes gradient */
6433: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6434: }
6435: if(mobilavproj > 0 )
6436: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6437: else
6438: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6439: for(i=1;i<=nlstate;i++){
6440: gp[i] = bprlim[i][i];
6441: mgp[theta][i] = bprlim[i][i];
6442: }
6443: for(i=1; i<=npar; i++) /* Computes gradient */
6444: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6445: if(mobilavproj > 0 )
6446: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6447: else
6448: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6449: for(i=1;i<=nlstate;i++){
6450: gm[i] = bprlim[i][i];
6451: mgm[theta][i] = bprlim[i][i];
6452: }
6453: for(i=1;i<=nlstate;i++)
6454: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6455: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6456: } /* End theta */
6457:
6458: trgradg =matrix(1,nlstate,1,npar);
6459:
6460: for(j=1; j<=nlstate;j++)
6461: for(theta=1; theta <=npar; theta++)
6462: trgradg[j][theta]=gradg[theta][j];
6463: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6464: /* printf("\nmgm mgp %d ",(int)age); */
6465: /* for(j=1; j<=nlstate;j++){ */
6466: /* printf(" %d ",j); */
6467: /* for(theta=1; theta <=npar; theta++) */
6468: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6469: /* printf("\n "); */
6470: /* } */
6471: /* } */
6472: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6473: /* printf("\n gradg %d ",(int)age); */
6474: /* for(j=1; j<=nlstate;j++){ */
6475: /* printf("%d ",j); */
6476: /* for(theta=1; theta <=npar; theta++) */
6477: /* printf("%d %lf ",theta,gradg[theta][j]); */
6478: /* printf("\n "); */
6479: /* } */
6480: /* } */
6481:
6482: for(i=1;i<=nlstate;i++)
6483: varbpl[i][(int)age] =0.;
6484: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6485: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6486: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6487: }else{
6488: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6489: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6490: }
6491: for(i=1;i<=nlstate;i++)
6492: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6493:
6494: fprintf(ficresvbl,"%.0f ",age );
6495: if(nresult >=1)
6496: fprintf(ficresvbl,"%d ",nres );
6497: for(i=1; i<=nlstate;i++)
6498: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6499: fprintf(ficresvbl,"\n");
6500: free_vector(gp,1,nlstate);
6501: free_vector(gm,1,nlstate);
6502: free_matrix(mgm,1,npar,1,nlstate);
6503: free_matrix(mgp,1,npar,1,nlstate);
6504: free_matrix(gradg,1,npar,1,nlstate);
6505: free_matrix(trgradg,1,nlstate,1,npar);
6506: } /* End age */
6507:
6508: free_vector(xp,1,npar);
6509: free_matrix(doldm,1,nlstate,1,npar);
6510: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6511:
6512: }
6513:
6514: /************ Variance of one-step probabilities ******************/
6515: 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 6516: {
6517: int i, j=0, k1, l1, tj;
6518: int k2, l2, j1, z1;
6519: int k=0, l;
6520: int first=1, first1, first2;
6521: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6522: double **dnewm,**doldm;
6523: double *xp;
6524: double *gp, *gm;
6525: double **gradg, **trgradg;
6526: double **mu;
6527: double age, cov[NCOVMAX+1];
6528: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6529: int theta;
6530: char fileresprob[FILENAMELENGTH];
6531: char fileresprobcov[FILENAMELENGTH];
6532: char fileresprobcor[FILENAMELENGTH];
6533: double ***varpij;
6534:
6535: strcpy(fileresprob,"PROB_");
6536: strcat(fileresprob,fileres);
6537: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6538: printf("Problem with resultfile: %s\n", fileresprob);
6539: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6540: }
6541: strcpy(fileresprobcov,"PROBCOV_");
6542: strcat(fileresprobcov,fileresu);
6543: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6544: printf("Problem with resultfile: %s\n", fileresprobcov);
6545: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6546: }
6547: strcpy(fileresprobcor,"PROBCOR_");
6548: strcat(fileresprobcor,fileresu);
6549: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6550: printf("Problem with resultfile: %s\n", fileresprobcor);
6551: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6552: }
6553: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6554: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6555: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6556: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6557: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6558: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6559: pstamp(ficresprob);
6560: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6561: fprintf(ficresprob,"# Age");
6562: pstamp(ficresprobcov);
6563: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6564: fprintf(ficresprobcov,"# Age");
6565: pstamp(ficresprobcor);
6566: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6567: fprintf(ficresprobcor,"# Age");
1.126 brouard 6568:
6569:
1.222 brouard 6570: for(i=1; i<=nlstate;i++)
6571: for(j=1; j<=(nlstate+ndeath);j++){
6572: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6573: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6574: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6575: }
6576: /* fprintf(ficresprob,"\n");
6577: fprintf(ficresprobcov,"\n");
6578: fprintf(ficresprobcor,"\n");
6579: */
6580: xp=vector(1,npar);
6581: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6582: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6583: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6584: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6585: first=1;
6586: fprintf(ficgp,"\n# Routine varprob");
6587: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6588: fprintf(fichtm,"\n");
6589:
1.288 brouard 6590: 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 6591: 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);
6592: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6593: and drawn. It helps understanding how is the covariance between two incidences.\
6594: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6595: 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 6596: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6597: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6598: standard deviations wide on each axis. <br>\
6599: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6600: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6601: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6602:
1.222 brouard 6603: cov[1]=1;
6604: /* tj=cptcoveff; */
1.225 brouard 6605: tj = (int) pow(2,cptcoveff);
1.222 brouard 6606: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6607: j1=0;
1.224 brouard 6608: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6609: if (cptcovn>0) {
6610: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6611: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6612: fprintf(ficresprob, "**********\n#\n");
6613: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6614: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6615: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6616:
1.222 brouard 6617: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6618: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6619: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6620:
6621:
1.222 brouard 6622: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6623: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6624: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6625:
1.222 brouard 6626: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6627: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6628: fprintf(ficresprobcor, "**********\n#");
6629: if(invalidvarcomb[j1]){
6630: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6631: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6632: continue;
6633: }
6634: }
6635: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6636: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6637: gp=vector(1,(nlstate)*(nlstate+ndeath));
6638: gm=vector(1,(nlstate)*(nlstate+ndeath));
6639: for (age=bage; age<=fage; age ++){
6640: cov[2]=age;
6641: if(nagesqr==1)
6642: cov[3]= age*age;
6643: for (k=1; k<=cptcovn;k++) {
6644: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6645: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6646: * 1 1 1 1 1
6647: * 2 2 1 1 1
6648: * 3 1 2 1 1
6649: */
6650: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6651: }
6652: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6653: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6654: for (k=1; k<=cptcovprod;k++)
6655: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6656:
6657:
1.222 brouard 6658: for(theta=1; theta <=npar; theta++){
6659: for(i=1; i<=npar; i++)
6660: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6661:
1.222 brouard 6662: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6663:
1.222 brouard 6664: k=0;
6665: for(i=1; i<= (nlstate); i++){
6666: for(j=1; j<=(nlstate+ndeath);j++){
6667: k=k+1;
6668: gp[k]=pmmij[i][j];
6669: }
6670: }
1.220 brouard 6671:
1.222 brouard 6672: for(i=1; i<=npar; i++)
6673: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6674:
1.222 brouard 6675: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6676: k=0;
6677: for(i=1; i<=(nlstate); i++){
6678: for(j=1; j<=(nlstate+ndeath);j++){
6679: k=k+1;
6680: gm[k]=pmmij[i][j];
6681: }
6682: }
1.220 brouard 6683:
1.222 brouard 6684: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6685: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6686: }
1.126 brouard 6687:
1.222 brouard 6688: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6689: for(theta=1; theta <=npar; theta++)
6690: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6691:
1.222 brouard 6692: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6693: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6694:
1.222 brouard 6695: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6696:
1.222 brouard 6697: k=0;
6698: for(i=1; i<=(nlstate); i++){
6699: for(j=1; j<=(nlstate+ndeath);j++){
6700: k=k+1;
6701: mu[k][(int) age]=pmmij[i][j];
6702: }
6703: }
6704: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6705: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6706: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6707:
1.222 brouard 6708: /*printf("\n%d ",(int)age);
6709: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6710: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6711: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6712: }*/
1.220 brouard 6713:
1.222 brouard 6714: fprintf(ficresprob,"\n%d ",(int)age);
6715: fprintf(ficresprobcov,"\n%d ",(int)age);
6716: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6717:
1.222 brouard 6718: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6719: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6720: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6721: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6722: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6723: }
6724: i=0;
6725: for (k=1; k<=(nlstate);k++){
6726: for (l=1; l<=(nlstate+ndeath);l++){
6727: i++;
6728: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6729: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6730: for (j=1; j<=i;j++){
6731: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6732: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6733: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6734: }
6735: }
6736: }/* end of loop for state */
6737: } /* end of loop for age */
6738: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6739: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6740: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6741: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6742:
6743: /* Confidence intervalle of pij */
6744: /*
6745: fprintf(ficgp,"\nunset parametric;unset label");
6746: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6747: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6748: 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);
6749: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6750: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6751: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6752: */
6753:
6754: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6755: first1=1;first2=2;
6756: for (k2=1; k2<=(nlstate);k2++){
6757: for (l2=1; l2<=(nlstate+ndeath);l2++){
6758: if(l2==k2) continue;
6759: j=(k2-1)*(nlstate+ndeath)+l2;
6760: for (k1=1; k1<=(nlstate);k1++){
6761: for (l1=1; l1<=(nlstate+ndeath);l1++){
6762: if(l1==k1) continue;
6763: i=(k1-1)*(nlstate+ndeath)+l1;
6764: if(i<=j) continue;
6765: for (age=bage; age<=fage; age ++){
6766: if ((int)age %5==0){
6767: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6768: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6769: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6770: mu1=mu[i][(int) age]/stepm*YEARM ;
6771: mu2=mu[j][(int) age]/stepm*YEARM;
6772: c12=cv12/sqrt(v1*v2);
6773: /* Computing eigen value of matrix of covariance */
6774: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6775: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6776: if ((lc2 <0) || (lc1 <0) ){
6777: if(first2==1){
6778: first1=0;
6779: 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);
6780: }
6781: 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);
6782: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6783: /* lc2=fabs(lc2); */
6784: }
1.220 brouard 6785:
1.222 brouard 6786: /* Eigen vectors */
1.280 brouard 6787: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6788: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6789: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6790: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6791: }else
6792: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6793: /*v21=sqrt(1.-v11*v11); *//* error */
6794: v21=(lc1-v1)/cv12*v11;
6795: v12=-v21;
6796: v22=v11;
6797: tnalp=v21/v11;
6798: if(first1==1){
6799: first1=0;
6800: 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);
6801: }
6802: 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);
6803: /*printf(fignu*/
6804: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6805: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6806: if(first==1){
6807: first=0;
6808: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6809: fprintf(ficgp,"\nset parametric;unset label");
6810: 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);
6811: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6812: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6813: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6814: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6815: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6816: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6817: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6818: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6819: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6820: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6821: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6822: 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 6823: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6824: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6825: }else{
6826: first=0;
6827: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6828: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6829: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6830: 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 6831: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6832: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6833: }/* if first */
6834: } /* age mod 5 */
6835: } /* end loop age */
6836: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6837: first=1;
6838: } /*l12 */
6839: } /* k12 */
6840: } /*l1 */
6841: }/* k1 */
6842: } /* loop on combination of covariates j1 */
6843: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6844: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6845: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6846: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6847: free_vector(xp,1,npar);
6848: fclose(ficresprob);
6849: fclose(ficresprobcov);
6850: fclose(ficresprobcor);
6851: fflush(ficgp);
6852: fflush(fichtmcov);
6853: }
1.126 brouard 6854:
6855:
6856: /******************* Printing html file ***********/
1.201 brouard 6857: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6858: int lastpass, int stepm, int weightopt, char model[],\
6859: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6860: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6861: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6862: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6863: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6864:
6865: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6866: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6867: </ul>");
1.237 brouard 6868: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6869: </ul>", model);
1.214 brouard 6870: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6871: 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",
6872: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6873: 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 6874: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6875: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6876: fprintf(fichtm,"\
6877: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6878: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6879: fprintf(fichtm,"\
1.217 brouard 6880: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6881: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6882: fprintf(fichtm,"\
1.288 brouard 6883: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6884: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6885: fprintf(fichtm,"\
1.288 brouard 6886: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6887: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6888: fprintf(fichtm,"\
1.211 brouard 6889: - (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 6890: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6891: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6892: if(prevfcast==1){
6893: fprintf(fichtm,"\
6894: - Prevalence projections by age and states: \
1.201 brouard 6895: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6896: }
1.126 brouard 6897:
6898:
1.225 brouard 6899: m=pow(2,cptcoveff);
1.222 brouard 6900: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6901:
1.264 brouard 6902: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6903:
6904: jj1=0;
6905:
6906: fprintf(fichtm," \n<ul>");
6907: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6908: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6909: if(m != 1 && TKresult[nres]!= k1)
6910: continue;
6911: jj1++;
6912: if (cptcovn > 0) {
6913: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6914: for (cpt=1; cpt<=cptcoveff;cpt++){
6915: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6916: }
6917: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6918: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6919: }
6920: fprintf(fichtm,"\">");
6921:
6922: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6923: fprintf(fichtm,"************ Results for covariates");
6924: for (cpt=1; cpt<=cptcoveff;cpt++){
6925: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6926: }
6927: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6928: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6929: }
6930: if(invalidvarcomb[k1]){
6931: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6932: continue;
6933: }
6934: fprintf(fichtm,"</a></li>");
6935: } /* cptcovn >0 */
6936: }
6937: fprintf(fichtm," \n</ul>");
6938:
1.222 brouard 6939: jj1=0;
1.237 brouard 6940:
6941: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6942: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6943: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6944: continue;
1.220 brouard 6945:
1.222 brouard 6946: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6947: jj1++;
6948: if (cptcovn > 0) {
1.264 brouard 6949: fprintf(fichtm,"\n<p><a name=\"rescov");
6950: for (cpt=1; cpt<=cptcoveff;cpt++){
6951: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6952: }
6953: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6954: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6955: }
6956: fprintf(fichtm,"\"</a>");
6957:
1.222 brouard 6958: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6959: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6960: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6961: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6962: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6963: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6964: }
1.237 brouard 6965: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6966: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6967: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6968: }
6969:
1.230 brouard 6970: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6971: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6972: if(invalidvarcomb[k1]){
6973: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6974: printf("\nCombination (%d) ignored because no cases \n",k1);
6975: continue;
6976: }
6977: }
6978: /* aij, bij */
1.259 brouard 6979: 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 6980: <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 6981: /* Pij */
1.241 brouard 6982: 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> \
6983: <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 6984: /* Quasi-incidences */
6985: 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 6986: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6987: 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 6988: 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> \
6989: <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 6990: /* Survival functions (period) in state j */
6991: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6992: 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 6993: <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 6994: }
6995: /* State specific survival functions (period) */
6996: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6997: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
6998: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 6999: <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 7000: }
1.288 brouard 7001: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7002: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7003: 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> \
7004: <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 7005: }
1.296 brouard 7006: if(prevbcast==1){
1.288 brouard 7007: /* Backward prevalence in each health state */
1.222 brouard 7008: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7009: 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 7010: <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 7011: }
1.217 brouard 7012: }
1.222 brouard 7013: if(prevfcast==1){
1.288 brouard 7014: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7015: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7016: 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 7017: <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 7018: }
7019: }
1.296 brouard 7020: if(prevbcast==1){
1.268 brouard 7021: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7022: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7023: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7024: 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 \
7025: 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) \
7026: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7027: <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 7028: }
7029: }
1.220 brouard 7030:
1.222 brouard 7031: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7032: 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> \
7033: <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 7034: }
7035: /* } /\* end i1 *\/ */
7036: }/* End k1 */
7037: fprintf(fichtm,"</ul>");
1.126 brouard 7038:
1.222 brouard 7039: fprintf(fichtm,"\
1.126 brouard 7040: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7041: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7042: - 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 7043: But because parameters are usually highly correlated (a higher incidence of disability \
7044: and a higher incidence of recovery can give very close observed transition) it might \
7045: be very useful to look not only at linear confidence intervals estimated from the \
7046: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7047: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7048: covariance matrix of the one-step probabilities. \
7049: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7050:
1.222 brouard 7051: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7052: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7053: fprintf(fichtm,"\
1.126 brouard 7054: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7055: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7056:
1.222 brouard 7057: fprintf(fichtm,"\
1.126 brouard 7058: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7059: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7060: fprintf(fichtm,"\
1.126 brouard 7061: - 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): \
7062: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7063: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7064: fprintf(fichtm,"\
1.126 brouard 7065: - (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): \
7066: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7067: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7068: fprintf(fichtm,"\
1.288 brouard 7069: - 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 7070: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7071: fprintf(fichtm,"\
1.128 brouard 7072: - 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 7073: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7074: fprintf(fichtm,"\
1.288 brouard 7075: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7076: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7077:
7078: /* if(popforecast==1) fprintf(fichtm,"\n */
7079: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7080: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7081: /* <br>",fileres,fileres,fileres,fileres); */
7082: /* else */
7083: /* 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 7084: fflush(fichtm);
7085: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7086:
1.225 brouard 7087: m=pow(2,cptcoveff);
1.222 brouard 7088: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7089:
1.222 brouard 7090: jj1=0;
1.237 brouard 7091:
1.241 brouard 7092: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7093: for(k1=1; k1<=m;k1++){
1.253 brouard 7094: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7095: continue;
1.222 brouard 7096: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7097: jj1++;
1.126 brouard 7098: if (cptcovn > 0) {
7099: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7100: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7101: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7102: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7103: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7104: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7105: }
7106:
1.126 brouard 7107: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7108:
1.222 brouard 7109: if(invalidvarcomb[k1]){
7110: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7111: continue;
7112: }
1.126 brouard 7113: }
7114: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7115: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7116: 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 7117: <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 7118: }
7119: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7120: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7121: true period expectancies (those weighted with period prevalences are also\
7122: drawn in addition to the population based expectancies computed using\
1.241 brouard 7123: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7124: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7125: /* } /\* end i1 *\/ */
7126: }/* End k1 */
1.241 brouard 7127: }/* End nres */
1.222 brouard 7128: fprintf(fichtm,"</ul>");
7129: fflush(fichtm);
1.126 brouard 7130: }
7131:
7132: /******************* Gnuplot file **************/
1.296 brouard 7133: 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 7134:
7135: char dirfileres[132],optfileres[132];
1.264 brouard 7136: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7137: 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 7138: int lv=0, vlv=0, kl=0;
1.130 brouard 7139: int ng=0;
1.201 brouard 7140: int vpopbased;
1.223 brouard 7141: int ioffset; /* variable offset for columns */
1.270 brouard 7142: int iyearc=1; /* variable column for year of projection */
7143: int iagec=1; /* variable column for age of projection */
1.235 brouard 7144: int nres=0; /* Index of resultline */
1.266 brouard 7145: int istart=1; /* For starting graphs in projections */
1.219 brouard 7146:
1.126 brouard 7147: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7148: /* printf("Problem with file %s",optionfilegnuplot); */
7149: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7150: /* } */
7151:
7152: /*#ifdef windows */
7153: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7154: /*#endif */
1.225 brouard 7155: m=pow(2,cptcoveff);
1.126 brouard 7156:
1.274 brouard 7157: /* diagram of the model */
7158: fprintf(ficgp,"\n#Diagram of the model \n");
7159: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7160: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7161: 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);
7162:
7163: 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);
7164: fprintf(ficgp,"\n#show arrow\nunset label\n");
7165: 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);
7166: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7167: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7168: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7169: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7170:
1.202 brouard 7171: /* Contribution to likelihood */
7172: /* Plot the probability implied in the likelihood */
1.223 brouard 7173: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7174: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7175: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7176: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7177: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7178: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7179: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7180: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7181: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7182: 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));
7183: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7184: 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));
7185: for (i=1; i<= nlstate ; i ++) {
7186: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7187: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7188: 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);
7189: for (j=2; j<= nlstate+ndeath ; j ++) {
7190: 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);
7191: }
7192: fprintf(ficgp,";\nset out; unset ylabel;\n");
7193: }
7194: /* 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 */
7195: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7196: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7197: fprintf(ficgp,"\nset out;unset log\n");
7198: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7199:
1.126 brouard 7200: strcpy(dirfileres,optionfilefiname);
7201: strcpy(optfileres,"vpl");
1.223 brouard 7202: /* 1eme*/
1.238 brouard 7203: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7204: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7205: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7206: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7207: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7208: continue;
7209: /* We are interested in selected combination by the resultline */
1.246 brouard 7210: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7211: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7212: strcpy(gplotlabel,"(");
1.238 brouard 7213: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7214: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7215: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7216: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7217: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7218: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7219: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7220: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7221: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7222: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7223: }
7224: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7225: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7226: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7227: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7228: }
7229: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7230: /* printf("\n#\n"); */
1.238 brouard 7231: fprintf(ficgp,"\n#\n");
7232: if(invalidvarcomb[k1]){
1.260 brouard 7233: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7234: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7235: continue;
7236: }
1.235 brouard 7237:
1.241 brouard 7238: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7239: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7240: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7241: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7242: 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);
7243: /* 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); */
7244: /* k1-1 error should be nres-1*/
1.238 brouard 7245: for (i=1; i<= nlstate ; i ++) {
7246: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7247: else fprintf(ficgp," %%*lf (%%*lf)");
7248: }
1.288 brouard 7249: 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 7250: for (i=1; i<= nlstate ; i ++) {
7251: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7252: else fprintf(ficgp," %%*lf (%%*lf)");
7253: }
1.260 brouard 7254: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7255: for (i=1; i<= nlstate ; i ++) {
7256: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7257: else fprintf(ficgp," %%*lf (%%*lf)");
7258: }
1.265 brouard 7259: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7260:
7261: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7262: if(cptcoveff ==0){
1.271 brouard 7263: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7264: }else{
7265: kl=0;
7266: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7267: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7268: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7269: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7270: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7271: vlv= nbcode[Tvaraff[k]][lv];
7272: kl++;
7273: /* 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 *\/ */
7274: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7275: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7276: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7277: if(k==cptcoveff){
7278: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7279: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7280: }else{
7281: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7282: kl++;
7283: }
7284: } /* end covariate */
7285: } /* end if no covariate */
7286:
1.296 brouard 7287: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7288: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7289: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7290: if(cptcoveff ==0){
1.245 brouard 7291: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7292: }else{
7293: kl=0;
7294: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7295: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7296: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7297: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7298: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7299: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7300: kl++;
1.238 brouard 7301: /* 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 *\/ */
7302: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7303: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7304: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7305: if(k==cptcoveff){
1.245 brouard 7306: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7307: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7308: }else{
7309: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7310: kl++;
7311: }
7312: } /* end covariate */
7313: } /* end if no covariate */
1.296 brouard 7314: if(prevbcast == 1){
1.268 brouard 7315: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7316: /* k1-1 error should be nres-1*/
7317: for (i=1; i<= nlstate ; i ++) {
7318: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7319: else fprintf(ficgp," %%*lf (%%*lf)");
7320: }
1.271 brouard 7321: 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 7322: for (i=1; i<= nlstate ; i ++) {
7323: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7324: else fprintf(ficgp," %%*lf (%%*lf)");
7325: }
1.276 brouard 7326: 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 7327: for (i=1; i<= nlstate ; i ++) {
7328: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7329: else fprintf(ficgp," %%*lf (%%*lf)");
7330: }
1.274 brouard 7331: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7332: } /* end if backprojcast */
1.296 brouard 7333: } /* end if prevbcast */
1.276 brouard 7334: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7335: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7336: } /* nres */
1.201 brouard 7337: } /* k1 */
7338: } /* cpt */
1.235 brouard 7339:
7340:
1.126 brouard 7341: /*2 eme*/
1.238 brouard 7342: for (k1=1; k1<= m ; k1 ++){
7343: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7344: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7345: continue;
7346: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7347: strcpy(gplotlabel,"(");
1.238 brouard 7348: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7349: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7350: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7351: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7352: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7353: vlv= nbcode[Tvaraff[k]][lv];
7354: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7355: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7356: }
1.237 brouard 7357: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7358: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7359: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7360: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7361: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7362: }
1.264 brouard 7363: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7364: fprintf(ficgp,"\n#\n");
1.223 brouard 7365: if(invalidvarcomb[k1]){
7366: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7367: continue;
7368: }
1.219 brouard 7369:
1.241 brouard 7370: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7371: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7372: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7373: if(vpopbased==0){
1.238 brouard 7374: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7375: }else
1.238 brouard 7376: fprintf(ficgp,"\nreplot ");
7377: for (i=1; i<= nlstate+1 ; i ++) {
7378: k=2*i;
1.261 brouard 7379: 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 7380: for (j=1; j<= nlstate+1 ; j ++) {
7381: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7382: else fprintf(ficgp," %%*lf (%%*lf)");
7383: }
7384: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7385: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7386: 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 7387: for (j=1; j<= nlstate+1 ; j ++) {
7388: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7389: else fprintf(ficgp," %%*lf (%%*lf)");
7390: }
7391: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7392: 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 7393: for (j=1; j<= nlstate+1 ; j ++) {
7394: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7395: else fprintf(ficgp," %%*lf (%%*lf)");
7396: }
7397: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7398: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7399: } /* state */
7400: } /* vpopbased */
1.264 brouard 7401: 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 7402: } /* end nres */
7403: } /* k1 end 2 eme*/
7404:
7405:
7406: /*3eme*/
7407: for (k1=1; k1<= m ; k1 ++){
7408: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7409: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7410: continue;
7411:
7412: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7413: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7414: strcpy(gplotlabel,"(");
1.238 brouard 7415: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7416: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7417: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7418: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7419: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7420: vlv= nbcode[Tvaraff[k]][lv];
7421: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7422: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7423: }
7424: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7425: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7426: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7427: }
1.264 brouard 7428: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7429: fprintf(ficgp,"\n#\n");
7430: if(invalidvarcomb[k1]){
7431: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7432: continue;
7433: }
7434:
7435: /* k=2+nlstate*(2*cpt-2); */
7436: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7437: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7438: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7439: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7440: 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 7441: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7442: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7443: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7444: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%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);
1.219 brouard 7447:
1.238 brouard 7448: */
7449: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7450: 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 7451: /* 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 7452:
1.238 brouard 7453: }
1.261 brouard 7454: 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 7455: }
1.264 brouard 7456: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7457: } /* end nres */
7458: } /* end kl 3eme */
1.126 brouard 7459:
1.223 brouard 7460: /* 4eme */
1.201 brouard 7461: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7462: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7463: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7464: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7465: continue;
1.238 brouard 7466: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7467: strcpy(gplotlabel,"(");
1.238 brouard 7468: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7469: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7470: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7471: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7472: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7473: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7474: vlv= nbcode[Tvaraff[k]][lv];
7475: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7476: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7477: }
7478: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7479: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7480: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7481: }
1.264 brouard 7482: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7483: fprintf(ficgp,"\n#\n");
7484: if(invalidvarcomb[k1]){
7485: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7486: continue;
1.223 brouard 7487: }
1.238 brouard 7488:
1.241 brouard 7489: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7490: 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 7491: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7492: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7493: k=3;
7494: for (i=1; i<= nlstate ; i ++){
7495: if(i==1){
7496: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7497: }else{
7498: fprintf(ficgp,", '' ");
7499: }
7500: l=(nlstate+ndeath)*(i-1)+1;
7501: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7502: for (j=2; j<= nlstate+ndeath ; j ++)
7503: fprintf(ficgp,"+$%d",k+l+j-1);
7504: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7505: } /* nlstate */
1.264 brouard 7506: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7507: } /* end cpt state*/
7508: } /* end nres */
7509: } /* end covariate k1 */
7510:
1.220 brouard 7511: /* 5eme */
1.201 brouard 7512: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7513: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7514: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7515: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7516: continue;
1.238 brouard 7517: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7518: strcpy(gplotlabel,"(");
1.238 brouard 7519: 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);
7520: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7521: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7522: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7523: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7524: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7525: vlv= nbcode[Tvaraff[k]][lv];
7526: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7527: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7528: }
7529: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7530: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7531: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7532: }
1.264 brouard 7533: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7534: fprintf(ficgp,"\n#\n");
7535: if(invalidvarcomb[k1]){
7536: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7537: continue;
7538: }
1.227 brouard 7539:
1.241 brouard 7540: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7541: 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 7542: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7543: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7544: k=3;
7545: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7546: if(j==1)
7547: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7548: else
7549: fprintf(ficgp,", '' ");
7550: l=(nlstate+ndeath)*(cpt-1) +j;
7551: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7552: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7553: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7554: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7555: } /* nlstate */
7556: fprintf(ficgp,", '' ");
7557: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7558: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7559: l=(nlstate+ndeath)*(cpt-1) +j;
7560: if(j < nlstate)
7561: fprintf(ficgp,"$%d +",k+l);
7562: else
7563: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7564: }
1.264 brouard 7565: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7566: } /* end cpt state*/
7567: } /* end covariate */
7568: } /* end nres */
1.227 brouard 7569:
1.220 brouard 7570: /* 6eme */
1.202 brouard 7571: /* CV preval stable (period) for each covariate */
1.237 brouard 7572: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7573: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7574: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7575: continue;
1.255 brouard 7576: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7577: strcpy(gplotlabel,"(");
1.288 brouard 7578: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7579: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7580: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7581: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7582: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7583: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7584: vlv= nbcode[Tvaraff[k]][lv];
7585: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7586: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7587: }
1.237 brouard 7588: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7589: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7590: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7591: }
1.264 brouard 7592: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7593: fprintf(ficgp,"\n#\n");
1.223 brouard 7594: if(invalidvarcomb[k1]){
1.227 brouard 7595: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7596: continue;
1.223 brouard 7597: }
1.227 brouard 7598:
1.241 brouard 7599: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7600: 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 7601: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7602: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7603: k=3; /* Offset */
1.255 brouard 7604: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7605: if(i==1)
7606: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7607: else
7608: fprintf(ficgp,", '' ");
1.255 brouard 7609: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7610: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7611: for (j=2; j<= nlstate ; j ++)
7612: fprintf(ficgp,"+$%d",k+l+j-1);
7613: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7614: } /* nlstate */
1.264 brouard 7615: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7616: } /* end cpt state*/
7617: } /* end covariate */
1.227 brouard 7618:
7619:
1.220 brouard 7620: /* 7eme */
1.296 brouard 7621: if(prevbcast == 1){
1.288 brouard 7622: /* CV backward prevalence for each covariate */
1.237 brouard 7623: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7624: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7625: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7626: continue;
1.268 brouard 7627: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7628: strcpy(gplotlabel,"(");
1.288 brouard 7629: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7630: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7631: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7632: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7633: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7634: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7635: vlv= nbcode[Tvaraff[k]][lv];
7636: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7637: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7638: }
1.237 brouard 7639: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7640: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7641: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7642: }
1.264 brouard 7643: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7644: fprintf(ficgp,"\n#\n");
7645: if(invalidvarcomb[k1]){
7646: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7647: continue;
7648: }
7649:
1.241 brouard 7650: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7651: 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 7652: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7653: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7654: k=3; /* Offset */
1.268 brouard 7655: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7656: if(i==1)
7657: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7658: else
7659: fprintf(ficgp,", '' ");
7660: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7661: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7662: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7663: /* 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 7664: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7665: /* for (j=2; j<= nlstate ; j ++) */
7666: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7667: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7668: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7669: } /* nlstate */
1.264 brouard 7670: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7671: } /* end cpt state*/
7672: } /* end covariate */
1.296 brouard 7673: } /* End if prevbcast */
1.218 brouard 7674:
1.223 brouard 7675: /* 8eme */
1.218 brouard 7676: if(prevfcast==1){
1.288 brouard 7677: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7678:
1.237 brouard 7679: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7680: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7681: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7682: continue;
1.211 brouard 7683: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7684: strcpy(gplotlabel,"(");
1.288 brouard 7685: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7686: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7687: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7688: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7689: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7690: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7691: vlv= nbcode[Tvaraff[k]][lv];
7692: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7693: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7694: }
1.237 brouard 7695: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7696: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7697: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7698: }
1.264 brouard 7699: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7700: fprintf(ficgp,"\n#\n");
7701: if(invalidvarcomb[k1]){
7702: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7703: continue;
7704: }
7705:
7706: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7707: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7708: 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 7709: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7710: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7711:
7712: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7713: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7714: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7715: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7716: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7717: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7718: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7719: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7720: if(i==istart){
1.227 brouard 7721: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7722: }else{
7723: fprintf(ficgp,",\\\n '' ");
7724: }
7725: if(cptcoveff ==0){ /* No covariate */
7726: ioffset=2; /* Age is in 2 */
7727: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7728: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7729: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7730: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7731: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7732: if(i==nlstate+1){
1.270 brouard 7733: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7734: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7735: fprintf(ficgp,",\\\n '' ");
7736: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7737: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7738: offyear, \
1.268 brouard 7739: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7740: }else
1.227 brouard 7741: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7742: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7743: }else{ /* more than 2 covariates */
1.270 brouard 7744: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7745: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7746: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7747: iyearc=ioffset-1;
7748: iagec=ioffset;
1.227 brouard 7749: fprintf(ficgp," u %d:(",ioffset);
7750: kl=0;
7751: strcpy(gplotcondition,"(");
7752: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7753: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7754: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7755: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7756: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7757: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7758: kl++;
7759: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7760: kl++;
7761: if(k <cptcoveff && cptcoveff>1)
7762: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7763: }
7764: strcpy(gplotcondition+strlen(gplotcondition),")");
7765: /* 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 *\/ */
7766: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7767: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7768: /* '' 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*/
7769: if(i==nlstate+1){
1.270 brouard 7770: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7771: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7772: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7773: fprintf(ficgp," u %d:(",iagec);
7774: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7775: iyearc, iagec, offyear, \
7776: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7777: /* '' 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 7778: }else{
7779: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7780: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7781: }
7782: } /* end if covariate */
7783: } /* nlstate */
1.264 brouard 7784: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7785: } /* end cpt state*/
7786: } /* end covariate */
7787: } /* End if prevfcast */
1.227 brouard 7788:
1.296 brouard 7789: if(prevbcast==1){
1.268 brouard 7790: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7791:
7792: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7793: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7794: if(m != 1 && TKresult[nres]!= k1)
7795: continue;
7796: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7797: strcpy(gplotlabel,"(");
7798: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7799: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7800: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7801: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7802: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7803: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7804: vlv= nbcode[Tvaraff[k]][lv];
7805: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7806: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7807: }
7808: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7809: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7810: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7811: }
7812: strcpy(gplotlabel+strlen(gplotlabel),")");
7813: fprintf(ficgp,"\n#\n");
7814: if(invalidvarcomb[k1]){
7815: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7816: continue;
7817: }
7818:
7819: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7820: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7821: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7822: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7823: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7824:
7825: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7826: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7827: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7828: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7829: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7830: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7831: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7832: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7833: if(i==istart){
7834: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7835: }else{
7836: fprintf(ficgp,",\\\n '' ");
7837: }
7838: if(cptcoveff ==0){ /* No covariate */
7839: ioffset=2; /* Age is in 2 */
7840: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7841: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7842: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7843: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7844: fprintf(ficgp," u %d:(", ioffset);
7845: if(i==nlstate+1){
1.270 brouard 7846: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7847: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7848: fprintf(ficgp,",\\\n '' ");
7849: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7850: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7851: offbyear, \
7852: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7853: }else
7854: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7855: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7856: }else{ /* more than 2 covariates */
1.270 brouard 7857: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7858: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7859: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7860: iyearc=ioffset-1;
7861: iagec=ioffset;
1.268 brouard 7862: fprintf(ficgp," u %d:(",ioffset);
7863: kl=0;
7864: strcpy(gplotcondition,"(");
7865: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7866: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7867: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7868: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7869: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7870: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7871: kl++;
7872: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7873: kl++;
7874: if(k <cptcoveff && cptcoveff>1)
7875: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7876: }
7877: strcpy(gplotcondition+strlen(gplotcondition),")");
7878: /* 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 *\/ */
7879: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7880: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7881: /* '' 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*/
7882: if(i==nlstate+1){
1.270 brouard 7883: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7884: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7885: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7886: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7887: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7888: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7889: iyearc,iagec,offbyear, \
7890: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7891: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7892: }else{
7893: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7894: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7895: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7896: }
7897: } /* end if covariate */
7898: } /* nlstate */
7899: fprintf(ficgp,"\nset out; unset label;\n");
7900: } /* end cpt state*/
7901: } /* end covariate */
1.296 brouard 7902: } /* End if prevbcast */
1.268 brouard 7903:
1.227 brouard 7904:
1.238 brouard 7905: /* 9eme writing MLE parameters */
7906: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7907: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7908: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7909: for(k=1; k <=(nlstate+ndeath); k++){
7910: if (k != i) {
1.227 brouard 7911: fprintf(ficgp,"# current state %d\n",k);
7912: for(j=1; j <=ncovmodel; j++){
7913: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7914: jk++;
7915: }
7916: fprintf(ficgp,"\n");
1.126 brouard 7917: }
7918: }
1.223 brouard 7919: }
1.187 brouard 7920: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7921:
1.145 brouard 7922: /*goto avoid;*/
1.238 brouard 7923: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7924: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7925: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7926: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7927: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7928: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7929: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7930: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7931: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7932: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7933: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7934: fprintf(ficgp,"# (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,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7937: fprintf(ficgp,"#\n");
1.223 brouard 7938: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7939: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7940: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7941: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7942: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7943: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7944: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7945: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7946: continue;
1.264 brouard 7947: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7948: strcpy(gplotlabel,"(");
1.276 brouard 7949: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7950: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7951: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7952: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7953: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7954: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7955: vlv= nbcode[Tvaraff[k]][lv];
7956: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7957: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7958: }
1.237 brouard 7959: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7960: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7961: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7962: }
1.264 brouard 7963: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7964: fprintf(ficgp,"\n#\n");
1.264 brouard 7965: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7966: fprintf(ficgp,"\nset key outside ");
7967: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7968: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7969: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7970: if (ng==1){
7971: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7972: fprintf(ficgp,"\nunset log y");
7973: }else if (ng==2){
7974: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7975: fprintf(ficgp,"\nset log y");
7976: }else if (ng==3){
7977: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7978: fprintf(ficgp,"\nset log y");
7979: }else
7980: fprintf(ficgp,"\nunset title ");
7981: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7982: i=1;
7983: for(k2=1; k2<=nlstate; k2++) {
7984: k3=i;
7985: for(k=1; k<=(nlstate+ndeath); k++) {
7986: if (k != k2){
7987: switch( ng) {
7988: case 1:
7989: if(nagesqr==0)
7990: fprintf(ficgp," p%d+p%d*x",i,i+1);
7991: else /* nagesqr =1 */
7992: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7993: break;
7994: case 2: /* ng=2 */
7995: if(nagesqr==0)
7996: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7997: else /* nagesqr =1 */
7998: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7999: break;
8000: case 3:
8001: if(nagesqr==0)
8002: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8003: else /* nagesqr =1 */
8004: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8005: break;
8006: }
8007: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8008: ijp=1; /* product no age */
8009: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8010: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8011: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8012: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8013: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8014: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8015: if(DummyV[j]==0){
8016: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8017: }else{ /* quantitative */
8018: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8019: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8020: }
8021: ij++;
1.237 brouard 8022: }
1.268 brouard 8023: }
8024: }else if(cptcovprod >0){
8025: if(j==Tprod[ijp]) { /* */
8026: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8027: if(ijp <=cptcovprod) { /* Product */
8028: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8029: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8030: /* 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)]); */
8031: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8032: }else{ /* Vn is dummy and Vm is quanti */
8033: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8034: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8035: }
8036: }else{ /* Vn*Vm Vn is quanti */
8037: if(DummyV[Tvard[ijp][2]]==0){
8038: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8039: }else{ /* Both quanti */
8040: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8041: }
1.237 brouard 8042: }
1.268 brouard 8043: ijp++;
1.237 brouard 8044: }
1.268 brouard 8045: } /* end Tprod */
1.237 brouard 8046: } else{ /* simple covariate */
1.264 brouard 8047: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8048: if(Dummy[j]==0){
8049: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8050: }else{ /* quantitative */
8051: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8052: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8053: }
1.237 brouard 8054: } /* end simple */
8055: } /* end j */
1.223 brouard 8056: }else{
8057: i=i-ncovmodel;
8058: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8059: fprintf(ficgp," (1.");
8060: }
1.227 brouard 8061:
1.223 brouard 8062: if(ng != 1){
8063: fprintf(ficgp,")/(1");
1.227 brouard 8064:
1.264 brouard 8065: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8066: if(nagesqr==0)
1.264 brouard 8067: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8068: else /* nagesqr =1 */
1.264 brouard 8069: 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 8070:
1.223 brouard 8071: ij=1;
8072: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8073: if(cptcovage >0){
8074: if((j-2)==Tage[ij]) { /* Bug valgrind */
8075: if(ij <=cptcovage) { /* Bug valgrind */
8076: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8077: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8078: ij++;
8079: }
8080: }
8081: }else
8082: 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 8083: }
8084: fprintf(ficgp,")");
8085: }
8086: fprintf(ficgp,")");
8087: if(ng ==2)
1.276 brouard 8088: 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 8089: else /* ng= 3 */
1.276 brouard 8090: 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 8091: }else{ /* end ng <> 1 */
8092: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8093: 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 8094: }
8095: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8096: fprintf(ficgp,",");
8097: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8098: fprintf(ficgp,",");
8099: i=i+ncovmodel;
8100: } /* end k */
8101: } /* end k2 */
1.276 brouard 8102: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8103: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8104: } /* end k1 */
1.223 brouard 8105: } /* end ng */
8106: /* avoid: */
8107: fflush(ficgp);
1.126 brouard 8108: } /* end gnuplot */
8109:
8110:
8111: /*************** Moving average **************/
1.219 brouard 8112: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8113: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8114:
1.222 brouard 8115: int i, cpt, cptcod;
8116: int modcovmax =1;
8117: int mobilavrange, mob;
8118: int iage=0;
1.288 brouard 8119: int firstA1=0, firstA2=0;
1.222 brouard 8120:
1.266 brouard 8121: double sum=0., sumr=0.;
1.222 brouard 8122: double age;
1.266 brouard 8123: double *sumnewp, *sumnewm, *sumnewmr;
8124: double *agemingood, *agemaxgood;
8125: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8126:
8127:
1.278 brouard 8128: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8129: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8130:
8131: sumnewp = vector(1,ncovcombmax);
8132: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8133: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8134: agemingood = vector(1,ncovcombmax);
1.266 brouard 8135: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8136: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8137: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8138:
8139: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8140: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8141: sumnewp[cptcod]=0.;
1.266 brouard 8142: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8143: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8144: }
8145: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8146:
1.266 brouard 8147: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8148: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8149: else mobilavrange=mobilav;
8150: for (age=bage; age<=fage; age++)
8151: for (i=1; i<=nlstate;i++)
8152: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8153: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8154: /* We keep the original values on the extreme ages bage, fage and for
8155: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8156: we use a 5 terms etc. until the borders are no more concerned.
8157: */
8158: for (mob=3;mob <=mobilavrange;mob=mob+2){
8159: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8160: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8161: sumnewm[cptcod]=0.;
8162: for (i=1; i<=nlstate;i++){
1.222 brouard 8163: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8164: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8165: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8166: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8167: }
8168: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8169: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8170: } /* end i */
8171: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8172: } /* end cptcod */
1.222 brouard 8173: }/* end age */
8174: }/* end mob */
1.266 brouard 8175: }else{
8176: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8177: return -1;
1.266 brouard 8178: }
8179:
8180: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8181: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8182: if(invalidvarcomb[cptcod]){
8183: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8184: continue;
8185: }
1.219 brouard 8186:
1.266 brouard 8187: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8188: sumnewm[cptcod]=0.;
8189: sumnewmr[cptcod]=0.;
8190: for (i=1; i<=nlstate;i++){
8191: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8192: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8193: }
8194: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8195: agemingoodr[cptcod]=age;
8196: }
8197: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8198: agemingood[cptcod]=age;
8199: }
8200: } /* age */
8201: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8202: sumnewm[cptcod]=0.;
1.266 brouard 8203: sumnewmr[cptcod]=0.;
1.222 brouard 8204: for (i=1; i<=nlstate;i++){
8205: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8206: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8207: }
8208: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8209: agemaxgoodr[cptcod]=age;
1.222 brouard 8210: }
8211: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8212: agemaxgood[cptcod]=age;
8213: }
8214: } /* age */
8215: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8216: /* but they will change */
1.288 brouard 8217: firstA1=0;firstA2=0;
1.266 brouard 8218: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8219: sumnewm[cptcod]=0.;
8220: sumnewmr[cptcod]=0.;
8221: for (i=1; i<=nlstate;i++){
8222: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8223: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8224: }
8225: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8226: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8227: agemaxgoodr[cptcod]=age; /* age min */
8228: for (i=1; i<=nlstate;i++)
8229: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8230: }else{ /* bad we change the value with the values of good ages */
8231: for (i=1; i<=nlstate;i++){
8232: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8233: } /* i */
8234: } /* end bad */
8235: }else{
8236: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8237: agemaxgood[cptcod]=age;
8238: }else{ /* bad we change the value with the values of good ages */
8239: for (i=1; i<=nlstate;i++){
8240: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8241: } /* i */
8242: } /* end bad */
8243: }/* end else */
8244: sum=0.;sumr=0.;
8245: for (i=1; i<=nlstate;i++){
8246: sum+=mobaverage[(int)age][i][cptcod];
8247: sumr+=probs[(int)age][i][cptcod];
8248: }
8249: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8250: if(!firstA1){
8251: firstA1=1;
8252: 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);
8253: }
8254: 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 8255: } /* end bad */
8256: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8257: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8258: if(!firstA2){
8259: firstA2=1;
8260: 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);
8261: }
8262: 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 8263: } /* end bad */
8264: }/* age */
1.266 brouard 8265:
8266: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8267: sumnewm[cptcod]=0.;
1.266 brouard 8268: sumnewmr[cptcod]=0.;
1.222 brouard 8269: for (i=1; i<=nlstate;i++){
8270: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8271: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8272: }
8273: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8274: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8275: agemingoodr[cptcod]=age;
8276: for (i=1; i<=nlstate;i++)
8277: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8278: }else{ /* bad we change the value with the values of good ages */
8279: for (i=1; i<=nlstate;i++){
8280: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8281: } /* i */
8282: } /* end bad */
8283: }else{
8284: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8285: agemingood[cptcod]=age;
8286: }else{ /* bad */
8287: for (i=1; i<=nlstate;i++){
8288: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8289: } /* i */
8290: } /* end bad */
8291: }/* end else */
8292: sum=0.;sumr=0.;
8293: for (i=1; i<=nlstate;i++){
8294: sum+=mobaverage[(int)age][i][cptcod];
8295: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8296: }
1.266 brouard 8297: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8298: 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 8299: } /* end bad */
8300: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8301: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8302: 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 8303: } /* end bad */
8304: }/* age */
1.266 brouard 8305:
1.222 brouard 8306:
8307: for (age=bage; age<=fage; age++){
1.235 brouard 8308: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8309: sumnewp[cptcod]=0.;
8310: sumnewm[cptcod]=0.;
8311: for (i=1; i<=nlstate;i++){
8312: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8313: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8314: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8315: }
8316: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8317: }
8318: /* printf("\n"); */
8319: /* } */
1.266 brouard 8320:
1.222 brouard 8321: /* brutal averaging */
1.266 brouard 8322: /* for (i=1; i<=nlstate;i++){ */
8323: /* for (age=1; age<=bage; age++){ */
8324: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8325: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8326: /* } */
8327: /* for (age=fage; age<=AGESUP; age++){ */
8328: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8329: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8330: /* } */
8331: /* } /\* end i status *\/ */
8332: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8333: /* for (age=1; age<=AGESUP; age++){ */
8334: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8335: /* mobaverage[(int)age][i][cptcod]=0.; */
8336: /* } */
8337: /* } */
1.222 brouard 8338: }/* end cptcod */
1.266 brouard 8339: free_vector(agemaxgoodr,1, ncovcombmax);
8340: free_vector(agemaxgood,1, ncovcombmax);
8341: free_vector(agemingood,1, ncovcombmax);
8342: free_vector(agemingoodr,1, ncovcombmax);
8343: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8344: free_vector(sumnewm,1, ncovcombmax);
8345: free_vector(sumnewp,1, ncovcombmax);
8346: return 0;
8347: }/* End movingaverage */
1.218 brouard 8348:
1.126 brouard 8349:
1.296 brouard 8350:
1.126 brouard 8351: /************** Forecasting ******************/
1.296 brouard 8352: /* 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)*/
8353: 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){
8354: /* dateintemean, mean date of interviews
8355: dateprojd, year, month, day of starting projection
8356: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8357: agemin, agemax range of age
8358: dateprev1 dateprev2 range of dates during which prevalence is computed
8359: */
1.296 brouard 8360: /* double anprojd, mprojd, jprojd; */
8361: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8362: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8363: double agec; /* generic age */
1.296 brouard 8364: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8365: double *popeffectif,*popcount;
8366: double ***p3mat;
1.218 brouard 8367: /* double ***mobaverage; */
1.126 brouard 8368: char fileresf[FILENAMELENGTH];
8369:
8370: agelim=AGESUP;
1.211 brouard 8371: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8372: in each health status at the date of interview (if between dateprev1 and dateprev2).
8373: We still use firstpass and lastpass as another selection.
8374: */
1.214 brouard 8375: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8376: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8377:
1.201 brouard 8378: strcpy(fileresf,"F_");
8379: strcat(fileresf,fileresu);
1.126 brouard 8380: if((ficresf=fopen(fileresf,"w"))==NULL) {
8381: printf("Problem with forecast resultfile: %s\n", fileresf);
8382: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8383: }
1.235 brouard 8384: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8385: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8386:
1.225 brouard 8387: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8388:
8389:
8390: stepsize=(int) (stepm+YEARM-1)/YEARM;
8391: if (stepm<=12) stepsize=1;
8392: if(estepm < stepm){
8393: printf ("Problem %d lower than %d\n",estepm, stepm);
8394: }
1.270 brouard 8395: else{
8396: hstepm=estepm;
8397: }
8398: if(estepm > stepm){ /* Yes every two year */
8399: stepsize=2;
8400: }
1.296 brouard 8401: hstepm=hstepm/stepm;
1.126 brouard 8402:
1.296 brouard 8403:
8404: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8405: /* fractional in yp1 *\/ */
8406: /* aintmean=yp; */
8407: /* yp2=modf((yp1*12),&yp); */
8408: /* mintmean=yp; */
8409: /* yp1=modf((yp2*30.5),&yp); */
8410: /* jintmean=yp; */
8411: /* if(jintmean==0) jintmean=1; */
8412: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8413:
1.296 brouard 8414:
8415: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8416: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8417: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8418: i1=pow(2,cptcoveff);
1.126 brouard 8419: if (cptcovn < 1){i1=1;}
8420:
1.296 brouard 8421: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8422:
8423: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8424:
1.126 brouard 8425: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8426: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8427: for(k=1; k<=i1;k++){
1.253 brouard 8428: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8429: continue;
1.227 brouard 8430: if(invalidvarcomb[k]){
8431: printf("\nCombination (%d) projection ignored because no cases \n",k);
8432: continue;
8433: }
8434: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8435: for(j=1;j<=cptcoveff;j++) {
8436: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8437: }
1.235 brouard 8438: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8439: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8440: }
1.227 brouard 8441: fprintf(ficresf," yearproj age");
8442: for(j=1; j<=nlstate+ndeath;j++){
8443: for(i=1; i<=nlstate;i++)
8444: fprintf(ficresf," p%d%d",i,j);
8445: fprintf(ficresf," wp.%d",j);
8446: }
1.296 brouard 8447: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8448: fprintf(ficresf,"\n");
1.296 brouard 8449: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8450: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8451: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8452: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8453: nhstepm = nhstepm/hstepm;
8454: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8455: oldm=oldms;savm=savms;
1.268 brouard 8456: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8457: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8458: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8459: for (h=0; h<=nhstepm; h++){
8460: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8461: break;
8462: }
8463: }
8464: fprintf(ficresf,"\n");
8465: for(j=1;j<=cptcoveff;j++)
8466: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8467: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8468:
8469: for(j=1; j<=nlstate+ndeath;j++) {
8470: ppij=0.;
8471: for(i=1; i<=nlstate;i++) {
1.278 brouard 8472: if (mobilav>=1)
8473: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8474: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8475: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8476: }
1.268 brouard 8477: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8478: } /* end i */
8479: fprintf(ficresf," %.3f", ppij);
8480: }/* end j */
1.227 brouard 8481: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8482: } /* end agec */
1.266 brouard 8483: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8484: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8485: } /* end yearp */
8486: } /* end k */
1.219 brouard 8487:
1.126 brouard 8488: fclose(ficresf);
1.215 brouard 8489: printf("End of Computing forecasting \n");
8490: fprintf(ficlog,"End of Computing forecasting\n");
8491:
1.126 brouard 8492: }
8493:
1.269 brouard 8494: /************** Back Forecasting ******************/
1.296 brouard 8495: /* 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){ */
8496: 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){
8497: /* back1, year, month, day of starting backprojection
1.267 brouard 8498: agemin, agemax range of age
8499: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8500: anback2 year of end of backprojection (same day and month as back1).
8501: prevacurrent and prev are prevalences.
1.267 brouard 8502: */
8503: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8504: double agec; /* generic age */
1.296 brouard 8505: double agelim, ppij, ppi, yp,yp1,yp2,jintmean,mintmean,aintmean;
1.267 brouard 8506: double *popeffectif,*popcount;
8507: double ***p3mat;
8508: /* double ***mobaverage; */
8509: char fileresfb[FILENAMELENGTH];
8510:
1.268 brouard 8511: agelim=AGEINF;
1.267 brouard 8512: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8513: in each health status at the date of interview (if between dateprev1 and dateprev2).
8514: We still use firstpass and lastpass as another selection.
8515: */
8516: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8517: /* firstpass, lastpass, stepm, weightopt, model); */
8518:
8519: /*Do we need to compute prevalence again?*/
8520:
8521: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8522:
8523: strcpy(fileresfb,"FB_");
8524: strcat(fileresfb,fileresu);
8525: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8526: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8527: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8528: }
8529: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8530: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8531:
8532: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8533:
8534:
8535: stepsize=(int) (stepm+YEARM-1)/YEARM;
8536: if (stepm<=12) stepsize=1;
8537: if(estepm < stepm){
8538: printf ("Problem %d lower than %d\n",estepm, stepm);
8539: }
1.270 brouard 8540: else{
8541: hstepm=estepm;
8542: }
8543: if(estepm >= stepm){ /* Yes every two year */
8544: stepsize=2;
8545: }
1.267 brouard 8546:
8547: hstepm=hstepm/stepm;
1.296 brouard 8548: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8549: /* fractional in yp1 *\/ */
8550: /* aintmean=yp; */
8551: /* yp2=modf((yp1*12),&yp); */
8552: /* mintmean=yp; */
8553: /* yp1=modf((yp2*30.5),&yp); */
8554: /* jintmean=yp; */
8555: /* if(jintmean==0) jintmean=1; */
8556: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8557:
8558: i1=pow(2,cptcoveff);
8559: if (cptcovn < 1){i1=1;}
8560:
1.296 brouard 8561: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8562: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8563:
8564: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8565:
8566: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8567: for(k=1; k<=i1;k++){
8568: if(i1 != 1 && TKresult[nres]!= k)
8569: continue;
8570: if(invalidvarcomb[k]){
8571: printf("\nCombination (%d) projection ignored because no cases \n",k);
8572: continue;
8573: }
1.268 brouard 8574: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8575: for(j=1;j<=cptcoveff;j++) {
8576: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8577: }
8578: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8579: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8580: }
8581: fprintf(ficresfb," yearbproj age");
8582: for(j=1; j<=nlstate+ndeath;j++){
8583: for(i=1; i<=nlstate;i++)
1.268 brouard 8584: fprintf(ficresfb," b%d%d",i,j);
8585: fprintf(ficresfb," b.%d",j);
1.267 brouard 8586: }
1.296 brouard 8587: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8588: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8589: fprintf(ficresfb,"\n");
1.296 brouard 8590: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8591: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8592: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8593: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8594: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8595: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8596: nhstepm = nhstepm/hstepm;
8597: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8598: oldm=oldms;savm=savms;
1.268 brouard 8599: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8600: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8601: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8602: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8603: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8604: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8605: for (h=0; h<=nhstepm; h++){
1.268 brouard 8606: if (h*hstepm/YEARM*stepm ==-yearp) {
8607: break;
8608: }
8609: }
8610: fprintf(ficresfb,"\n");
8611: for(j=1;j<=cptcoveff;j++)
8612: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8613: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8614: for(i=1; i<=nlstate+ndeath;i++) {
8615: ppij=0.;ppi=0.;
8616: for(j=1; j<=nlstate;j++) {
8617: /* if (mobilav==1) */
1.269 brouard 8618: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8619: ppi=ppi+prevacurrent[(int)agec][j][k];
8620: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8621: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8622: /* else { */
8623: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8624: /* } */
1.268 brouard 8625: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8626: } /* end j */
8627: if(ppi <0.99){
8628: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8629: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8630: }
8631: fprintf(ficresfb," %.3f", ppij);
8632: }/* end j */
1.267 brouard 8633: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8634: } /* end agec */
8635: } /* end yearp */
8636: } /* end k */
1.217 brouard 8637:
1.267 brouard 8638: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8639:
1.267 brouard 8640: fclose(ficresfb);
8641: printf("End of Computing Back forecasting \n");
8642: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8643:
1.267 brouard 8644: }
1.217 brouard 8645:
1.269 brouard 8646: /* Variance of prevalence limit: varprlim */
8647: 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 8648: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8649:
8650: char fileresvpl[FILENAMELENGTH];
8651: FILE *ficresvpl;
8652: double **oldm, **savm;
8653: double **varpl; /* Variances of prevalence limits by age */
8654: int i1, k, nres, j ;
8655:
8656: strcpy(fileresvpl,"VPL_");
8657: strcat(fileresvpl,fileresu);
8658: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8659: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8660: exit(0);
8661: }
1.288 brouard 8662: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8663: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8664:
8665: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8666: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8667:
8668: i1=pow(2,cptcoveff);
8669: if (cptcovn < 1){i1=1;}
8670:
8671: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8672: for(k=1; k<=i1;k++){
8673: if(i1 != 1 && TKresult[nres]!= k)
8674: continue;
8675: fprintf(ficresvpl,"\n#****** ");
8676: printf("\n#****** ");
8677: fprintf(ficlog,"\n#****** ");
8678: for(j=1;j<=cptcoveff;j++) {
8679: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8680: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8681: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8682: }
8683: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8684: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8685: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8686: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8687: }
8688: fprintf(ficresvpl,"******\n");
8689: printf("******\n");
8690: fprintf(ficlog,"******\n");
8691:
8692: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8693: oldm=oldms;savm=savms;
8694: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8695: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8696: /*}*/
8697: }
8698:
8699: fclose(ficresvpl);
1.288 brouard 8700: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8701: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8702:
8703: }
8704: /* Variance of back prevalence: varbprlim */
8705: 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){
8706: /*------- Variance of back (stable) prevalence------*/
8707:
8708: char fileresvbl[FILENAMELENGTH];
8709: FILE *ficresvbl;
8710:
8711: double **oldm, **savm;
8712: double **varbpl; /* Variances of back prevalence limits by age */
8713: int i1, k, nres, j ;
8714:
8715: strcpy(fileresvbl,"VBL_");
8716: strcat(fileresvbl,fileresu);
8717: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8718: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8719: exit(0);
8720: }
8721: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8722: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8723:
8724:
8725: i1=pow(2,cptcoveff);
8726: if (cptcovn < 1){i1=1;}
8727:
8728: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8729: for(k=1; k<=i1;k++){
8730: if(i1 != 1 && TKresult[nres]!= k)
8731: continue;
8732: fprintf(ficresvbl,"\n#****** ");
8733: printf("\n#****** ");
8734: fprintf(ficlog,"\n#****** ");
8735: for(j=1;j<=cptcoveff;j++) {
8736: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8737: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8738: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8739: }
8740: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8741: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8742: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8743: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8744: }
8745: fprintf(ficresvbl,"******\n");
8746: printf("******\n");
8747: fprintf(ficlog,"******\n");
8748:
8749: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8750: oldm=oldms;savm=savms;
8751:
8752: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8753: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8754: /*}*/
8755: }
8756:
8757: fclose(ficresvbl);
8758: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8759: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8760:
8761: } /* End of varbprlim */
8762:
1.126 brouard 8763: /************** Forecasting *****not tested NB*************/
1.227 brouard 8764: /* 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 8765:
1.227 brouard 8766: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8767: /* int *popage; */
8768: /* double calagedatem, agelim, kk1, kk2; */
8769: /* double *popeffectif,*popcount; */
8770: /* double ***p3mat,***tabpop,***tabpopprev; */
8771: /* /\* double ***mobaverage; *\/ */
8772: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8773:
1.227 brouard 8774: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8775: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8776: /* agelim=AGESUP; */
8777: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8778:
1.227 brouard 8779: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8780:
8781:
1.227 brouard 8782: /* strcpy(filerespop,"POP_"); */
8783: /* strcat(filerespop,fileresu); */
8784: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8785: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8786: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8787: /* } */
8788: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8789: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8790:
1.227 brouard 8791: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8792:
1.227 brouard 8793: /* /\* if (mobilav!=0) { *\/ */
8794: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8795: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8796: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8797: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8798: /* /\* } *\/ */
8799: /* /\* } *\/ */
1.126 brouard 8800:
1.227 brouard 8801: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8802: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8803:
1.227 brouard 8804: /* agelim=AGESUP; */
1.126 brouard 8805:
1.227 brouard 8806: /* hstepm=1; */
8807: /* hstepm=hstepm/stepm; */
1.218 brouard 8808:
1.227 brouard 8809: /* if (popforecast==1) { */
8810: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8811: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8812: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8813: /* } */
8814: /* popage=ivector(0,AGESUP); */
8815: /* popeffectif=vector(0,AGESUP); */
8816: /* popcount=vector(0,AGESUP); */
1.126 brouard 8817:
1.227 brouard 8818: /* i=1; */
8819: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8820:
1.227 brouard 8821: /* imx=i; */
8822: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8823: /* } */
1.218 brouard 8824:
1.227 brouard 8825: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8826: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8827: /* k=k+1; */
8828: /* fprintf(ficrespop,"\n#******"); */
8829: /* for(j=1;j<=cptcoveff;j++) { */
8830: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8831: /* } */
8832: /* fprintf(ficrespop,"******\n"); */
8833: /* fprintf(ficrespop,"# Age"); */
8834: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8835: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8836:
1.227 brouard 8837: /* for (cpt=0; cpt<=0;cpt++) { */
8838: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8839:
1.227 brouard 8840: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8841: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8842: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8843:
1.227 brouard 8844: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8845: /* oldm=oldms;savm=savms; */
8846: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8847:
1.227 brouard 8848: /* for (h=0; h<=nhstepm; h++){ */
8849: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8850: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8851: /* } */
8852: /* for(j=1; j<=nlstate+ndeath;j++) { */
8853: /* kk1=0.;kk2=0; */
8854: /* for(i=1; i<=nlstate;i++) { */
8855: /* if (mobilav==1) */
8856: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8857: /* else { */
8858: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8859: /* } */
8860: /* } */
8861: /* if (h==(int)(calagedatem+12*cpt)){ */
8862: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8863: /* /\*fprintf(ficrespop," %.3f", kk1); */
8864: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8865: /* } */
8866: /* } */
8867: /* for(i=1; i<=nlstate;i++){ */
8868: /* kk1=0.; */
8869: /* for(j=1; j<=nlstate;j++){ */
8870: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8871: /* } */
8872: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8873: /* } */
1.218 brouard 8874:
1.227 brouard 8875: /* if (h==(int)(calagedatem+12*cpt)) */
8876: /* for(j=1; j<=nlstate;j++) */
8877: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8878: /* } */
8879: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8880: /* } */
8881: /* } */
1.218 brouard 8882:
1.227 brouard 8883: /* /\******\/ */
1.218 brouard 8884:
1.227 brouard 8885: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8886: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8887: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8888: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8889: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8890:
1.227 brouard 8891: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8892: /* oldm=oldms;savm=savms; */
8893: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8894: /* for (h=0; h<=nhstepm; h++){ */
8895: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8896: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8897: /* } */
8898: /* for(j=1; j<=nlstate+ndeath;j++) { */
8899: /* kk1=0.;kk2=0; */
8900: /* for(i=1; i<=nlstate;i++) { */
8901: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8902: /* } */
8903: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8904: /* } */
8905: /* } */
8906: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8907: /* } */
8908: /* } */
8909: /* } */
8910: /* } */
1.218 brouard 8911:
1.227 brouard 8912: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8913:
1.227 brouard 8914: /* if (popforecast==1) { */
8915: /* free_ivector(popage,0,AGESUP); */
8916: /* free_vector(popeffectif,0,AGESUP); */
8917: /* free_vector(popcount,0,AGESUP); */
8918: /* } */
8919: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8920: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8921: /* fclose(ficrespop); */
8922: /* } /\* End of popforecast *\/ */
1.218 brouard 8923:
1.126 brouard 8924: int fileappend(FILE *fichier, char *optionfich)
8925: {
8926: if((fichier=fopen(optionfich,"a"))==NULL) {
8927: printf("Problem with file: %s\n", optionfich);
8928: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8929: return (0);
8930: }
8931: fflush(fichier);
8932: return (1);
8933: }
8934:
8935:
8936: /**************** function prwizard **********************/
8937: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8938: {
8939:
8940: /* Wizard to print covariance matrix template */
8941:
1.164 brouard 8942: char ca[32], cb[32];
8943: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8944: int numlinepar;
8945:
8946: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8947: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8948: for(i=1; i <=nlstate; i++){
8949: jj=0;
8950: for(j=1; j <=nlstate+ndeath; j++){
8951: if(j==i) continue;
8952: jj++;
8953: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8954: printf("%1d%1d",i,j);
8955: fprintf(ficparo,"%1d%1d",i,j);
8956: for(k=1; k<=ncovmodel;k++){
8957: /* printf(" %lf",param[i][j][k]); */
8958: /* fprintf(ficparo," %lf",param[i][j][k]); */
8959: printf(" 0.");
8960: fprintf(ficparo," 0.");
8961: }
8962: printf("\n");
8963: fprintf(ficparo,"\n");
8964: }
8965: }
8966: printf("# Scales (for hessian or gradient estimation)\n");
8967: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8968: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8969: for(i=1; i <=nlstate; i++){
8970: jj=0;
8971: for(j=1; j <=nlstate+ndeath; j++){
8972: if(j==i) continue;
8973: jj++;
8974: fprintf(ficparo,"%1d%1d",i,j);
8975: printf("%1d%1d",i,j);
8976: fflush(stdout);
8977: for(k=1; k<=ncovmodel;k++){
8978: /* printf(" %le",delti3[i][j][k]); */
8979: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8980: printf(" 0.");
8981: fprintf(ficparo," 0.");
8982: }
8983: numlinepar++;
8984: printf("\n");
8985: fprintf(ficparo,"\n");
8986: }
8987: }
8988: printf("# Covariance matrix\n");
8989: /* # 121 Var(a12)\n\ */
8990: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8991: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8992: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8993: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8994: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8995: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8996: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8997: fflush(stdout);
8998: fprintf(ficparo,"# Covariance matrix\n");
8999: /* # 121 Var(a12)\n\ */
9000: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9001: /* # ...\n\ */
9002: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9003:
9004: for(itimes=1;itimes<=2;itimes++){
9005: jj=0;
9006: for(i=1; i <=nlstate; i++){
9007: for(j=1; j <=nlstate+ndeath; j++){
9008: if(j==i) continue;
9009: for(k=1; k<=ncovmodel;k++){
9010: jj++;
9011: ca[0]= k+'a'-1;ca[1]='\0';
9012: if(itimes==1){
9013: printf("#%1d%1d%d",i,j,k);
9014: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9015: }else{
9016: printf("%1d%1d%d",i,j,k);
9017: fprintf(ficparo,"%1d%1d%d",i,j,k);
9018: /* printf(" %.5le",matcov[i][j]); */
9019: }
9020: ll=0;
9021: for(li=1;li <=nlstate; li++){
9022: for(lj=1;lj <=nlstate+ndeath; lj++){
9023: if(lj==li) continue;
9024: for(lk=1;lk<=ncovmodel;lk++){
9025: ll++;
9026: if(ll<=jj){
9027: cb[0]= lk +'a'-1;cb[1]='\0';
9028: if(ll<jj){
9029: if(itimes==1){
9030: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9031: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9032: }else{
9033: printf(" 0.");
9034: fprintf(ficparo," 0.");
9035: }
9036: }else{
9037: if(itimes==1){
9038: printf(" Var(%s%1d%1d)",ca,i,j);
9039: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9040: }else{
9041: printf(" 0.");
9042: fprintf(ficparo," 0.");
9043: }
9044: }
9045: }
9046: } /* end lk */
9047: } /* end lj */
9048: } /* end li */
9049: printf("\n");
9050: fprintf(ficparo,"\n");
9051: numlinepar++;
9052: } /* end k*/
9053: } /*end j */
9054: } /* end i */
9055: } /* end itimes */
9056:
9057: } /* end of prwizard */
9058: /******************* Gompertz Likelihood ******************************/
9059: double gompertz(double x[])
9060: {
9061: double A,B,L=0.0,sump=0.,num=0.;
9062: int i,n=0; /* n is the size of the sample */
9063:
1.220 brouard 9064: for (i=1;i<=imx ; i++) {
1.126 brouard 9065: sump=sump+weight[i];
9066: /* sump=sump+1;*/
9067: num=num+1;
9068: }
9069:
9070:
9071: /* for (i=0; i<=imx; i++)
9072: 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]);*/
9073:
9074: for (i=1;i<=imx ; i++)
9075: {
9076: if (cens[i] == 1 && wav[i]>1)
9077: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9078:
9079: if (cens[i] == 0 && wav[i]>1)
9080: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
9081: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
9082:
9083: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9084: if (wav[i] > 1 ) { /* ??? */
9085: L=L+A*weight[i];
9086: /* 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]);*/
9087: }
9088: }
9089:
9090: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9091:
9092: return -2*L*num/sump;
9093: }
9094:
1.136 brouard 9095: #ifdef GSL
9096: /******************* Gompertz_f Likelihood ******************************/
9097: double gompertz_f(const gsl_vector *v, void *params)
9098: {
9099: double A,B,LL=0.0,sump=0.,num=0.;
9100: double *x= (double *) v->data;
9101: int i,n=0; /* n is the size of the sample */
9102:
9103: for (i=0;i<=imx-1 ; i++) {
9104: sump=sump+weight[i];
9105: /* sump=sump+1;*/
9106: num=num+1;
9107: }
9108:
9109:
9110: /* for (i=0; i<=imx; i++)
9111: 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]);*/
9112: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9113: for (i=1;i<=imx ; i++)
9114: {
9115: if (cens[i] == 1 && wav[i]>1)
9116: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9117:
9118: if (cens[i] == 0 && wav[i]>1)
9119: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9120: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9121:
9122: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9123: if (wav[i] > 1 ) { /* ??? */
9124: LL=LL+A*weight[i];
9125: /* 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]);*/
9126: }
9127: }
9128:
9129: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9130: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9131:
9132: return -2*LL*num/sump;
9133: }
9134: #endif
9135:
1.126 brouard 9136: /******************* Printing html file ***********/
1.201 brouard 9137: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9138: int lastpass, int stepm, int weightopt, char model[],\
9139: int imx, double p[],double **matcov,double agemortsup){
9140: int i,k;
9141:
9142: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9143: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9144: for (i=1;i<=2;i++)
9145: 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 9146: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9147: fprintf(fichtm,"</ul>");
9148:
9149: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9150:
9151: 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>");
9152:
9153: for (k=agegomp;k<(agemortsup-2);k++)
9154: 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]);
9155:
9156:
9157: fflush(fichtm);
9158: }
9159:
9160: /******************* Gnuplot file **************/
1.201 brouard 9161: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9162:
9163: char dirfileres[132],optfileres[132];
1.164 brouard 9164:
1.126 brouard 9165: int ng;
9166:
9167:
9168: /*#ifdef windows */
9169: fprintf(ficgp,"cd \"%s\" \n",pathc);
9170: /*#endif */
9171:
9172:
9173: strcpy(dirfileres,optionfilefiname);
9174: strcpy(optfileres,"vpl");
1.199 brouard 9175: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9176: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9177: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9178: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9179: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9180:
9181: }
9182:
1.136 brouard 9183: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9184: {
1.126 brouard 9185:
1.136 brouard 9186: /*-------- data file ----------*/
9187: FILE *fic;
9188: char dummy[]=" ";
1.240 brouard 9189: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9190: int lstra;
1.136 brouard 9191: int linei, month, year,iout;
9192: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9193: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9194: char *stratrunc;
1.223 brouard 9195:
1.240 brouard 9196: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9197: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9198:
1.240 brouard 9199: for(v=1; v <=ncovcol;v++){
9200: DummyV[v]=0;
9201: FixedV[v]=0;
9202: }
9203: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9204: DummyV[v]=1;
9205: FixedV[v]=0;
9206: }
9207: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9208: DummyV[v]=0;
9209: FixedV[v]=1;
9210: }
9211: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9212: DummyV[v]=1;
9213: FixedV[v]=1;
9214: }
9215: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9216: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9217: 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]);
9218: }
1.126 brouard 9219:
1.136 brouard 9220: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9221: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9222: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9223: }
1.126 brouard 9224:
1.136 brouard 9225: i=1;
9226: linei=0;
9227: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9228: linei=linei+1;
9229: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9230: if(line[j] == '\t')
9231: line[j] = ' ';
9232: }
9233: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9234: ;
9235: };
9236: line[j+1]=0; /* Trims blanks at end of line */
9237: if(line[0]=='#'){
9238: fprintf(ficlog,"Comment line\n%s\n",line);
9239: printf("Comment line\n%s\n",line);
9240: continue;
9241: }
9242: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9243: strcpy(line, linetmp);
1.223 brouard 9244:
9245: /* Loops on waves */
9246: for (j=maxwav;j>=1;j--){
9247: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9248: cutv(stra, strb, line, ' ');
9249: if(strb[0]=='.') { /* Missing value */
9250: lval=-1;
9251: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9252: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9253: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9254: 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);
9255: 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);
9256: return 1;
9257: }
9258: }else{
9259: errno=0;
9260: /* what_kind_of_number(strb); */
9261: dval=strtod(strb,&endptr);
9262: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9263: /* if(strb != endptr && *endptr == '\0') */
9264: /* dval=dlval; */
9265: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9266: if( strb[0]=='\0' || (*endptr != '\0')){
9267: 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);
9268: 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);
9269: return 1;
9270: }
9271: cotqvar[j][iv][i]=dval;
9272: cotvar[j][ntv+iv][i]=dval;
9273: }
9274: strcpy(line,stra);
1.223 brouard 9275: }/* end loop ntqv */
1.225 brouard 9276:
1.223 brouard 9277: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9278: cutv(stra, strb, line, ' ');
9279: if(strb[0]=='.') { /* Missing value */
9280: lval=-1;
9281: }else{
9282: errno=0;
9283: lval=strtol(strb,&endptr,10);
9284: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9285: if( strb[0]=='\0' || (*endptr != '\0')){
9286: 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);
9287: 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);
9288: return 1;
9289: }
9290: }
9291: if(lval <-1 || lval >1){
9292: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9293: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9294: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9295: For example, for multinomial values like 1, 2 and 3,\n \
9296: build V1=0 V2=0 for the reference value (1),\n \
9297: V1=1 V2=0 for (2) \n \
1.223 brouard 9298: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9299: output of IMaCh is often meaningless.\n \
1.223 brouard 9300: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9301: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9302: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9303: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9304: For example, for multinomial values like 1, 2 and 3,\n \
9305: build V1=0 V2=0 for the reference value (1),\n \
9306: V1=1 V2=0 for (2) \n \
1.223 brouard 9307: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9308: output of IMaCh is often meaningless.\n \
1.223 brouard 9309: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9310: return 1;
9311: }
9312: cotvar[j][iv][i]=(double)(lval);
9313: strcpy(line,stra);
1.223 brouard 9314: }/* end loop ntv */
1.225 brouard 9315:
1.223 brouard 9316: /* Statuses at wave */
1.137 brouard 9317: cutv(stra, strb, line, ' ');
1.223 brouard 9318: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9319: lval=-1;
1.136 brouard 9320: }else{
1.238 brouard 9321: errno=0;
9322: lval=strtol(strb,&endptr,10);
9323: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9324: if( strb[0]=='\0' || (*endptr != '\0')){
9325: 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);
9326: 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);
9327: return 1;
9328: }
1.136 brouard 9329: }
1.225 brouard 9330:
1.136 brouard 9331: s[j][i]=lval;
1.225 brouard 9332:
1.223 brouard 9333: /* Date of Interview */
1.136 brouard 9334: strcpy(line,stra);
9335: cutv(stra, strb,line,' ');
1.169 brouard 9336: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9337: }
1.169 brouard 9338: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9339: month=99;
9340: year=9999;
1.136 brouard 9341: }else{
1.225 brouard 9342: 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);
9343: 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);
9344: return 1;
1.136 brouard 9345: }
9346: anint[j][i]= (double) year;
9347: mint[j][i]= (double)month;
9348: strcpy(line,stra);
1.223 brouard 9349: } /* End loop on waves */
1.225 brouard 9350:
1.223 brouard 9351: /* Date of death */
1.136 brouard 9352: cutv(stra, strb,line,' ');
1.169 brouard 9353: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9354: }
1.169 brouard 9355: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9356: month=99;
9357: year=9999;
9358: }else{
1.141 brouard 9359: 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 9360: 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);
9361: return 1;
1.136 brouard 9362: }
9363: andc[i]=(double) year;
9364: moisdc[i]=(double) month;
9365: strcpy(line,stra);
9366:
1.223 brouard 9367: /* Date of birth */
1.136 brouard 9368: cutv(stra, strb,line,' ');
1.169 brouard 9369: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9370: }
1.169 brouard 9371: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9372: month=99;
9373: year=9999;
9374: }else{
1.141 brouard 9375: 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);
9376: 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 9377: return 1;
1.136 brouard 9378: }
9379: if (year==9999) {
1.141 brouard 9380: 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);
9381: 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 9382: return 1;
9383:
1.136 brouard 9384: }
9385: annais[i]=(double)(year);
9386: moisnais[i]=(double)(month);
9387: strcpy(line,stra);
1.225 brouard 9388:
1.223 brouard 9389: /* Sample weight */
1.136 brouard 9390: cutv(stra, strb,line,' ');
9391: errno=0;
9392: dval=strtod(strb,&endptr);
9393: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9394: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9395: 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 9396: fflush(ficlog);
9397: return 1;
9398: }
9399: weight[i]=dval;
9400: strcpy(line,stra);
1.225 brouard 9401:
1.223 brouard 9402: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9403: cutv(stra, strb, line, ' ');
9404: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9405: lval=-1;
1.223 brouard 9406: }else{
1.225 brouard 9407: errno=0;
9408: /* what_kind_of_number(strb); */
9409: dval=strtod(strb,&endptr);
9410: /* if(strb != endptr && *endptr == '\0') */
9411: /* dval=dlval; */
9412: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9413: if( strb[0]=='\0' || (*endptr != '\0')){
9414: 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);
9415: 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);
9416: return 1;
9417: }
9418: coqvar[iv][i]=dval;
1.226 brouard 9419: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9420: }
9421: strcpy(line,stra);
9422: }/* end loop nqv */
1.136 brouard 9423:
1.223 brouard 9424: /* Covariate values */
1.136 brouard 9425: for (j=ncovcol;j>=1;j--){
9426: cutv(stra, strb,line,' ');
1.223 brouard 9427: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9428: lval=-1;
1.136 brouard 9429: }else{
1.225 brouard 9430: errno=0;
9431: lval=strtol(strb,&endptr,10);
9432: if( strb[0]=='\0' || (*endptr != '\0')){
9433: 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);
9434: 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);
9435: return 1;
9436: }
1.136 brouard 9437: }
9438: if(lval <-1 || lval >1){
1.225 brouard 9439: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9440: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9441: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9442: For example, for multinomial values like 1, 2 and 3,\n \
9443: build V1=0 V2=0 for the reference value (1),\n \
9444: V1=1 V2=0 for (2) \n \
1.136 brouard 9445: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9446: output of IMaCh is often meaningless.\n \
1.136 brouard 9447: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9448: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9449: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9450: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9451: For example, for multinomial values like 1, 2 and 3,\n \
9452: build V1=0 V2=0 for the reference value (1),\n \
9453: V1=1 V2=0 for (2) \n \
1.136 brouard 9454: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9455: output of IMaCh is often meaningless.\n \
1.136 brouard 9456: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9457: return 1;
1.136 brouard 9458: }
9459: covar[j][i]=(double)(lval);
9460: strcpy(line,stra);
9461: }
9462: lstra=strlen(stra);
1.225 brouard 9463:
1.136 brouard 9464: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9465: stratrunc = &(stra[lstra-9]);
9466: num[i]=atol(stratrunc);
9467: }
9468: else
9469: num[i]=atol(stra);
9470: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9471: 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;}*/
9472:
9473: i=i+1;
9474: } /* End loop reading data */
1.225 brouard 9475:
1.136 brouard 9476: *imax=i-1; /* Number of individuals */
9477: fclose(fic);
1.225 brouard 9478:
1.136 brouard 9479: return (0);
1.164 brouard 9480: /* endread: */
1.225 brouard 9481: printf("Exiting readdata: ");
9482: fclose(fic);
9483: return (1);
1.223 brouard 9484: }
1.126 brouard 9485:
1.234 brouard 9486: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9487: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9488: while (*p2 == ' ')
1.234 brouard 9489: p2++;
9490: /* while ((*p1++ = *p2++) !=0) */
9491: /* ; */
9492: /* do */
9493: /* while (*p2 == ' ') */
9494: /* p2++; */
9495: /* while (*p1++ == *p2++); */
9496: *stri=p2;
1.145 brouard 9497: }
9498:
1.235 brouard 9499: int decoderesult ( char resultline[], int nres)
1.230 brouard 9500: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9501: {
1.235 brouard 9502: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9503: char resultsav[MAXLINE];
1.234 brouard 9504: int resultmodel[MAXLINE];
9505: int modelresult[MAXLINE];
1.230 brouard 9506: char stra[80], strb[80], strc[80], strd[80],stre[80];
9507:
1.234 brouard 9508: removefirstspace(&resultline);
1.233 brouard 9509: printf("decoderesult:%s\n",resultline);
1.230 brouard 9510:
9511: if (strstr(resultline,"v") !=0){
9512: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9513: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9514: return 1;
9515: }
9516: trimbb(resultsav, resultline);
9517: if (strlen(resultsav) >1){
9518: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9519: }
1.253 brouard 9520: if(j == 0){ /* Resultline but no = */
9521: TKresult[nres]=0; /* Combination for the nresult and the model */
9522: return (0);
9523: }
9524:
1.234 brouard 9525: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9526: 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);
9527: 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);
9528: }
9529: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9530: if(nbocc(resultsav,'=') >1){
9531: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9532: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9533: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9534: }else
9535: cutl(strc,strd,resultsav,'=');
1.230 brouard 9536: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9537:
1.230 brouard 9538: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9539: Tvarsel[k]=atoi(strc);
9540: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9541: /* cptcovsel++; */
9542: if (nbocc(stra,'=') >0)
9543: strcpy(resultsav,stra); /* and analyzes it */
9544: }
1.235 brouard 9545: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9546: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9547: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9548: match=0;
1.236 brouard 9549: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9550: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9551: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9552: match=1;
9553: break;
9554: }
9555: }
9556: if(match == 0){
9557: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9558: }
9559: }
9560: }
1.235 brouard 9561: /* Checking for missing or useless values in comparison of current model needs */
9562: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9563: match=0;
1.235 brouard 9564: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9565: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9566: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9567: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9568: ++match;
9569: }
9570: }
9571: }
9572: if(match == 0){
9573: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9574: }else if(match > 1){
9575: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9576: }
9577: }
1.235 brouard 9578:
1.234 brouard 9579: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9580: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9581: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9582: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9583: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9584: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9585: /* 1 0 0 0 */
9586: /* 2 1 0 0 */
9587: /* 3 0 1 0 */
9588: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9589: /* 5 0 0 1 */
9590: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9591: /* 7 0 1 1 */
9592: /* 8 1 1 1 */
1.237 brouard 9593: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9594: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9595: /* V5*age V5 known which value for nres? */
9596: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9597: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9598: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9599: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9600: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9601: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9602: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9603: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9604: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9605: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9606: k4++;;
9607: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9608: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9609: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9610: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9611: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9612: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9613: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9614: k4q++;;
9615: }
9616: }
1.234 brouard 9617:
1.235 brouard 9618: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9619: return (0);
9620: }
1.235 brouard 9621:
1.230 brouard 9622: int decodemodel( char model[], int lastobs)
9623: /**< This routine decodes the model and returns:
1.224 brouard 9624: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9625: * - nagesqr = 1 if age*age in the model, otherwise 0.
9626: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9627: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9628: * - cptcovage number of covariates with age*products =2
9629: * - cptcovs number of simple covariates
9630: * - 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
9631: * which is a new column after the 9 (ncovcol) variables.
9632: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9633: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9634: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9635: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9636: */
1.136 brouard 9637: {
1.238 brouard 9638: int i, j, k, ks, v;
1.227 brouard 9639: int j1, k1, k2, k3, k4;
1.136 brouard 9640: char modelsav[80];
1.145 brouard 9641: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9642: char *strpt;
1.136 brouard 9643:
1.145 brouard 9644: /*removespace(model);*/
1.136 brouard 9645: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9646: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9647: if (strstr(model,"AGE") !=0){
1.192 brouard 9648: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9649: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9650: return 1;
9651: }
1.141 brouard 9652: if (strstr(model,"v") !=0){
9653: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9654: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9655: return 1;
9656: }
1.187 brouard 9657: strcpy(modelsav,model);
9658: if ((strpt=strstr(model,"age*age")) !=0){
9659: printf(" strpt=%s, model=%s\n",strpt, model);
9660: if(strpt != model){
1.234 brouard 9661: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9662: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9663: corresponding column of parameters.\n",model);
1.234 brouard 9664: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9665: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9666: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9667: return 1;
1.225 brouard 9668: }
1.187 brouard 9669: nagesqr=1;
9670: if (strstr(model,"+age*age") !=0)
1.234 brouard 9671: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9672: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9673: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9674: else
1.234 brouard 9675: substrchaine(modelsav, model, "age*age");
1.187 brouard 9676: }else
9677: nagesqr=0;
9678: if (strlen(modelsav) >1){
9679: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9680: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9681: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9682: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9683: * cst, age and age*age
9684: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9685: /* including age products which are counted in cptcovage.
9686: * but the covariates which are products must be treated
9687: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9688: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9689: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9690:
9691:
1.187 brouard 9692: /* Design
9693: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9694: * < ncovcol=8 >
9695: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9696: * k= 1 2 3 4 5 6 7 8
9697: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9698: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9699: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9700: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9701: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9702: * Tage[++cptcovage]=k
9703: * if products, new covar are created after ncovcol with k1
9704: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9705: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9706: * 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
9707: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9708: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9709: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9710: * < ncovcol=8 >
9711: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9712: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9713: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9714: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9715: * p Tprod[1]@2={ 6, 5}
9716: *p Tvard[1][1]@4= {7, 8, 5, 6}
9717: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9718: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9719: *How to reorganize?
9720: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9721: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9722: * {2, 1, 4, 8, 5, 6, 3, 7}
9723: * Struct []
9724: */
1.225 brouard 9725:
1.187 brouard 9726: /* This loop fills the array Tvar from the string 'model'.*/
9727: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9728: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9729: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9730: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9731: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9732: /* k=1 Tvar[1]=2 (from V2) */
9733: /* k=5 Tvar[5] */
9734: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9735: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9736: /* } */
1.198 brouard 9737: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9738: /*
9739: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9740: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9741: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9742: }
1.187 brouard 9743: cptcovage=0;
9744: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9745: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9746: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9747: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9748: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9749: /*scanf("%d",i);*/
9750: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9751: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9752: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9753: /* covar is not filled and then is empty */
9754: cptcovprod--;
9755: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9756: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9757: Typevar[k]=1; /* 1 for age product */
9758: cptcovage++; /* Sums the number of covariates which include age as a product */
9759: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9760: /*printf("stre=%s ", stre);*/
9761: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9762: cptcovprod--;
9763: cutl(stre,strb,strc,'V');
9764: Tvar[k]=atoi(stre);
9765: Typevar[k]=1; /* 1 for age product */
9766: cptcovage++;
9767: Tage[cptcovage]=k;
9768: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9769: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9770: cptcovn++;
9771: cptcovprodnoage++;k1++;
9772: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9773: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9774: because this model-covariate is a construction we invent a new column
9775: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9776: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9777: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9778: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9779: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9780: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9781: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9782: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9783: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9784: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9785: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9786: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9787: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9788: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9789: for (i=1; i<=lastobs;i++){
9790: /* Computes the new covariate which is a product of
9791: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9792: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9793: }
9794: } /* End age is not in the model */
9795: } /* End if model includes a product */
9796: else { /* no more sum */
9797: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9798: /* scanf("%d",i);*/
9799: cutl(strd,strc,strb,'V');
9800: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9801: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9802: Tvar[k]=atoi(strd);
9803: Typevar[k]=0; /* 0 for simple covariates */
9804: }
9805: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9806: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9807: scanf("%d",i);*/
1.187 brouard 9808: } /* end of loop + on total covariates */
9809: } /* end if strlen(modelsave == 0) age*age might exist */
9810: } /* end if strlen(model == 0) */
1.136 brouard 9811:
9812: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9813: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9814:
1.136 brouard 9815: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9816: printf("cptcovprod=%d ", cptcovprod);
9817: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9818: scanf("%d ",i);*/
9819:
9820:
1.230 brouard 9821: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9822: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9823: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9824: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9825: k = 1 2 3 4 5 6 7 8 9
9826: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9827: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9828: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9829: Dummy[k] 1 0 0 0 3 1 1 2 3
9830: Tmodelind[combination of covar]=k;
1.225 brouard 9831: */
9832: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9833: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9834: /* 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 9835: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9836: printf("Model=%s\n\
9837: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9838: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9839: 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);
9840: fprintf(ficlog,"Model=%s\n\
9841: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9842: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9843: 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 9844: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9845: 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 */
9846: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9847: Fixed[k]= 0;
9848: Dummy[k]= 0;
1.225 brouard 9849: ncoveff++;
1.232 brouard 9850: ncovf++;
1.234 brouard 9851: nsd++;
9852: modell[k].maintype= FTYPE;
9853: TvarsD[nsd]=Tvar[k];
9854: TvarsDind[nsd]=k;
9855: TvarF[ncovf]=Tvar[k];
9856: TvarFind[ncovf]=k;
9857: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9858: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9859: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9860: Fixed[k]= 0;
9861: Dummy[k]= 0;
9862: ncoveff++;
9863: ncovf++;
9864: modell[k].maintype= FTYPE;
9865: TvarF[ncovf]=Tvar[k];
9866: TvarFind[ncovf]=k;
1.230 brouard 9867: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9868: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9869: }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 9870: Fixed[k]= 0;
9871: Dummy[k]= 1;
1.230 brouard 9872: nqfveff++;
1.234 brouard 9873: modell[k].maintype= FTYPE;
9874: modell[k].subtype= FQ;
9875: nsq++;
9876: TvarsQ[nsq]=Tvar[k];
9877: TvarsQind[nsq]=k;
1.232 brouard 9878: ncovf++;
1.234 brouard 9879: TvarF[ncovf]=Tvar[k];
9880: TvarFind[ncovf]=k;
1.231 brouard 9881: 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 9882: 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 9883: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9884: Fixed[k]= 1;
9885: Dummy[k]= 0;
1.225 brouard 9886: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9887: modell[k].maintype= VTYPE;
9888: modell[k].subtype= VD;
9889: nsd++;
9890: TvarsD[nsd]=Tvar[k];
9891: TvarsDind[nsd]=k;
9892: ncovv++; /* Only simple time varying variables */
9893: TvarV[ncovv]=Tvar[k];
1.242 brouard 9894: 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 9895: 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 */
9896: 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 9897: 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);
9898: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9899: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9900: Fixed[k]= 1;
9901: Dummy[k]= 1;
9902: nqtveff++;
9903: modell[k].maintype= VTYPE;
9904: modell[k].subtype= VQ;
9905: ncovv++; /* Only simple time varying variables */
9906: nsq++;
9907: TvarsQ[nsq]=Tvar[k];
9908: TvarsQind[nsq]=k;
9909: TvarV[ncovv]=Tvar[k];
1.242 brouard 9910: 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 9911: 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 */
9912: 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 9913: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9914: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9915: 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 9916: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9917: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9918: ncova++;
9919: TvarA[ncova]=Tvar[k];
9920: TvarAind[ncova]=k;
1.231 brouard 9921: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9922: Fixed[k]= 2;
9923: Dummy[k]= 2;
9924: modell[k].maintype= ATYPE;
9925: modell[k].subtype= APFD;
9926: /* ncoveff++; */
1.227 brouard 9927: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9928: Fixed[k]= 2;
9929: Dummy[k]= 3;
9930: modell[k].maintype= ATYPE;
9931: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9932: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9933: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9934: Fixed[k]= 3;
9935: Dummy[k]= 2;
9936: modell[k].maintype= ATYPE;
9937: modell[k].subtype= APVD; /* Product age * varying dummy */
9938: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9939: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9940: Fixed[k]= 3;
9941: Dummy[k]= 3;
9942: modell[k].maintype= ATYPE;
9943: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9944: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9945: }
9946: }else if (Typevar[k] == 2) { /* product without age */
9947: k1=Tposprod[k];
9948: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9949: if(Tvard[k1][2] <=ncovcol){
9950: Fixed[k]= 1;
9951: Dummy[k]= 0;
9952: modell[k].maintype= FTYPE;
9953: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9954: ncovf++; /* Fixed variables without age */
9955: TvarF[ncovf]=Tvar[k];
9956: TvarFind[ncovf]=k;
9957: }else if(Tvard[k1][2] <=ncovcol+nqv){
9958: Fixed[k]= 0; /* or 2 ?*/
9959: Dummy[k]= 1;
9960: modell[k].maintype= FTYPE;
9961: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9962: ncovf++; /* Varying variables without age */
9963: TvarF[ncovf]=Tvar[k];
9964: TvarFind[ncovf]=k;
9965: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9966: Fixed[k]= 1;
9967: Dummy[k]= 0;
9968: modell[k].maintype= VTYPE;
9969: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9970: ncovv++; /* Varying variables without age */
9971: TvarV[ncovv]=Tvar[k];
9972: TvarVind[ncovv]=k;
9973: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9974: Fixed[k]= 1;
9975: Dummy[k]= 1;
9976: modell[k].maintype= VTYPE;
9977: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9978: ncovv++; /* Varying variables without age */
9979: TvarV[ncovv]=Tvar[k];
9980: TvarVind[ncovv]=k;
9981: }
1.227 brouard 9982: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9983: if(Tvard[k1][2] <=ncovcol){
9984: Fixed[k]= 0; /* or 2 ?*/
9985: Dummy[k]= 1;
9986: modell[k].maintype= FTYPE;
9987: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9988: ncovf++; /* Fixed variables without age */
9989: TvarF[ncovf]=Tvar[k];
9990: TvarFind[ncovf]=k;
9991: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9992: Fixed[k]= 1;
9993: Dummy[k]= 1;
9994: modell[k].maintype= VTYPE;
9995: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9996: ncovv++; /* Varying variables without age */
9997: TvarV[ncovv]=Tvar[k];
9998: TvarVind[ncovv]=k;
9999: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10000: Fixed[k]= 1;
10001: Dummy[k]= 1;
10002: modell[k].maintype= VTYPE;
10003: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10004: ncovv++; /* Varying variables without age */
10005: TvarV[ncovv]=Tvar[k];
10006: TvarVind[ncovv]=k;
10007: ncovv++; /* Varying variables without age */
10008: TvarV[ncovv]=Tvar[k];
10009: TvarVind[ncovv]=k;
10010: }
1.227 brouard 10011: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10012: if(Tvard[k1][2] <=ncovcol){
10013: Fixed[k]= 1;
10014: Dummy[k]= 1;
10015: modell[k].maintype= VTYPE;
10016: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10017: ncovv++; /* Varying variables without age */
10018: TvarV[ncovv]=Tvar[k];
10019: TvarVind[ncovv]=k;
10020: }else if(Tvard[k1][2] <=ncovcol+nqv){
10021: Fixed[k]= 1;
10022: Dummy[k]= 1;
10023: modell[k].maintype= VTYPE;
10024: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10025: ncovv++; /* Varying variables without age */
10026: TvarV[ncovv]=Tvar[k];
10027: TvarVind[ncovv]=k;
10028: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10029: Fixed[k]= 1;
10030: Dummy[k]= 0;
10031: modell[k].maintype= VTYPE;
10032: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10033: ncovv++; /* Varying variables without age */
10034: TvarV[ncovv]=Tvar[k];
10035: TvarVind[ncovv]=k;
10036: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10037: Fixed[k]= 1;
10038: Dummy[k]= 1;
10039: modell[k].maintype= VTYPE;
10040: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10041: ncovv++; /* Varying variables without age */
10042: TvarV[ncovv]=Tvar[k];
10043: TvarVind[ncovv]=k;
10044: }
1.227 brouard 10045: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10046: if(Tvard[k1][2] <=ncovcol){
10047: Fixed[k]= 1;
10048: Dummy[k]= 1;
10049: modell[k].maintype= VTYPE;
10050: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10051: ncovv++; /* Varying variables without age */
10052: TvarV[ncovv]=Tvar[k];
10053: TvarVind[ncovv]=k;
10054: }else if(Tvard[k1][2] <=ncovcol+nqv){
10055: Fixed[k]= 1;
10056: Dummy[k]= 1;
10057: modell[k].maintype= VTYPE;
10058: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10059: ncovv++; /* Varying variables without age */
10060: TvarV[ncovv]=Tvar[k];
10061: TvarVind[ncovv]=k;
10062: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10063: Fixed[k]= 1;
10064: Dummy[k]= 1;
10065: modell[k].maintype= VTYPE;
10066: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10067: ncovv++; /* Varying variables without age */
10068: TvarV[ncovv]=Tvar[k];
10069: TvarVind[ncovv]=k;
10070: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10071: Fixed[k]= 1;
10072: Dummy[k]= 1;
10073: modell[k].maintype= VTYPE;
10074: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10075: ncovv++; /* Varying variables without age */
10076: TvarV[ncovv]=Tvar[k];
10077: TvarVind[ncovv]=k;
10078: }
1.227 brouard 10079: }else{
1.240 brouard 10080: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10081: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10082: } /*end k1*/
1.225 brouard 10083: }else{
1.226 brouard 10084: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10085: 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 10086: }
1.227 brouard 10087: 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 10088: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10089: 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]);
10090: }
10091: /* Searching for doublons in the model */
10092: for(k1=1; k1<= cptcovt;k1++){
10093: for(k2=1; k2 <k1;k2++){
1.285 brouard 10094: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10095: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10096: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10097: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10098: 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]);
10099: 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 10100: return(1);
10101: }
10102: }else if (Typevar[k1] ==2){
10103: k3=Tposprod[k1];
10104: k4=Tposprod[k2];
10105: 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])) ){
10106: 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]]);
10107: 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);
10108: return(1);
10109: }
10110: }
1.227 brouard 10111: }
10112: }
1.225 brouard 10113: }
10114: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10115: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10116: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10117: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10118: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10119: /*endread:*/
1.225 brouard 10120: printf("Exiting decodemodel: ");
10121: return (1);
1.136 brouard 10122: }
10123:
1.169 brouard 10124: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10125: {/* Check ages at death */
1.136 brouard 10126: int i, m;
1.218 brouard 10127: int firstone=0;
10128:
1.136 brouard 10129: for (i=1; i<=imx; i++) {
10130: for(m=2; (m<= maxwav); m++) {
10131: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10132: anint[m][i]=9999;
1.216 brouard 10133: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10134: s[m][i]=-1;
1.136 brouard 10135: }
10136: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10137: *nberr = *nberr + 1;
1.218 brouard 10138: if(firstone == 0){
10139: firstone=1;
1.260 brouard 10140: 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 10141: }
1.262 brouard 10142: 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 10143: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10144: }
10145: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10146: (*nberr)++;
1.259 brouard 10147: 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 10148: 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 10149: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10150: }
10151: }
10152: }
10153:
10154: for (i=1; i<=imx; i++) {
10155: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10156: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10157: 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 10158: if (s[m][i] >= nlstate+1) {
1.169 brouard 10159: if(agedc[i]>0){
10160: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10161: agev[m][i]=agedc[i];
1.214 brouard 10162: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10163: }else {
1.136 brouard 10164: if ((int)andc[i]!=9999){
10165: nbwarn++;
10166: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10167: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10168: agev[m][i]=-1;
10169: }
10170: }
1.169 brouard 10171: } /* agedc > 0 */
1.214 brouard 10172: } /* end if */
1.136 brouard 10173: else if(s[m][i] !=9){ /* Standard case, age in fractional
10174: years but with the precision of a month */
10175: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10176: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10177: agev[m][i]=1;
10178: else if(agev[m][i] < *agemin){
10179: *agemin=agev[m][i];
10180: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10181: }
10182: else if(agev[m][i] >*agemax){
10183: *agemax=agev[m][i];
1.156 brouard 10184: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10185: }
10186: /*agev[m][i]=anint[m][i]-annais[i];*/
10187: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10188: } /* en if 9*/
1.136 brouard 10189: else { /* =9 */
1.214 brouard 10190: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10191: agev[m][i]=1;
10192: s[m][i]=-1;
10193: }
10194: }
1.214 brouard 10195: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10196: agev[m][i]=1;
1.214 brouard 10197: else{
10198: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10199: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10200: agev[m][i]=0;
10201: }
10202: } /* End for lastpass */
10203: }
1.136 brouard 10204:
10205: for (i=1; i<=imx; i++) {
10206: for(m=firstpass; (m<=lastpass); m++){
10207: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10208: (*nberr)++;
1.136 brouard 10209: 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);
10210: 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);
10211: return 1;
10212: }
10213: }
10214: }
10215:
10216: /*for (i=1; i<=imx; i++){
10217: for (m=firstpass; (m<lastpass); m++){
10218: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10219: }
10220:
10221: }*/
10222:
10223:
1.139 brouard 10224: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10225: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10226:
10227: return (0);
1.164 brouard 10228: /* endread:*/
1.136 brouard 10229: printf("Exiting calandcheckages: ");
10230: return (1);
10231: }
10232:
1.172 brouard 10233: #if defined(_MSC_VER)
10234: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10235: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10236: //#include "stdafx.h"
10237: //#include <stdio.h>
10238: //#include <tchar.h>
10239: //#include <windows.h>
10240: //#include <iostream>
10241: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10242:
10243: LPFN_ISWOW64PROCESS fnIsWow64Process;
10244:
10245: BOOL IsWow64()
10246: {
10247: BOOL bIsWow64 = FALSE;
10248:
10249: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10250: // (HANDLE, PBOOL);
10251:
10252: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10253:
10254: HMODULE module = GetModuleHandle(_T("kernel32"));
10255: const char funcName[] = "IsWow64Process";
10256: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10257: GetProcAddress(module, funcName);
10258:
10259: if (NULL != fnIsWow64Process)
10260: {
10261: if (!fnIsWow64Process(GetCurrentProcess(),
10262: &bIsWow64))
10263: //throw std::exception("Unknown error");
10264: printf("Unknown error\n");
10265: }
10266: return bIsWow64 != FALSE;
10267: }
10268: #endif
1.177 brouard 10269:
1.191 brouard 10270: void syscompilerinfo(int logged)
1.292 brouard 10271: {
10272: #include <stdint.h>
10273:
10274: /* #include "syscompilerinfo.h"*/
1.185 brouard 10275: /* command line Intel compiler 32bit windows, XP compatible:*/
10276: /* /GS /W3 /Gy
10277: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10278: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10279: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10280: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10281: */
10282: /* 64 bits */
1.185 brouard 10283: /*
10284: /GS /W3 /Gy
10285: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10286: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10287: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10288: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10289: /* Optimization are useless and O3 is slower than O2 */
10290: /*
10291: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10292: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10293: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10294: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10295: */
1.186 brouard 10296: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10297: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10298: /PDB:"visual studio
10299: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10300: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10301: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10302: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10303: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10304: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10305: uiAccess='false'"
10306: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10307: /NOLOGO /TLBID:1
10308: */
1.292 brouard 10309:
10310:
1.177 brouard 10311: #if defined __INTEL_COMPILER
1.178 brouard 10312: #if defined(__GNUC__)
10313: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10314: #endif
1.177 brouard 10315: #elif defined(__GNUC__)
1.179 brouard 10316: #ifndef __APPLE__
1.174 brouard 10317: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10318: #endif
1.177 brouard 10319: struct utsname sysInfo;
1.178 brouard 10320: int cross = CROSS;
10321: if (cross){
10322: printf("Cross-");
1.191 brouard 10323: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10324: }
1.174 brouard 10325: #endif
10326:
1.191 brouard 10327: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10328: #if defined(__clang__)
1.191 brouard 10329: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10330: #endif
10331: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10332: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10333: #endif
10334: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10335: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10336: #endif
10337: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10338: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10339: #endif
10340: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10341: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10342: #endif
10343: #if defined(_MSC_VER)
1.191 brouard 10344: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10345: #endif
10346: #if defined(__PGI)
1.191 brouard 10347: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10348: #endif
10349: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10350: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10351: #endif
1.191 brouard 10352: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10353:
1.167 brouard 10354: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10355: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10356: // Windows (x64 and x86)
1.191 brouard 10357: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10358: #elif __unix__ // all unices, not all compilers
10359: // Unix
1.191 brouard 10360: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10361: #elif __linux__
10362: // linux
1.191 brouard 10363: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10364: #elif __APPLE__
1.174 brouard 10365: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10366: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10367: #endif
10368:
10369: /* __MINGW32__ */
10370: /* __CYGWIN__ */
10371: /* __MINGW64__ */
10372: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10373: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10374: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10375: /* _WIN64 // Defined for applications for Win64. */
10376: /* _M_X64 // Defined for compilations that target x64 processors. */
10377: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10378:
1.167 brouard 10379: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10380: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10381: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10382: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10383: #else
1.191 brouard 10384: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10385: #endif
10386:
1.169 brouard 10387: #if defined(__GNUC__)
10388: # if defined(__GNUC_PATCHLEVEL__)
10389: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10390: + __GNUC_MINOR__ * 100 \
10391: + __GNUC_PATCHLEVEL__)
10392: # else
10393: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10394: + __GNUC_MINOR__ * 100)
10395: # endif
1.174 brouard 10396: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10397: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10398:
10399: if (uname(&sysInfo) != -1) {
10400: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10401: 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 10402: }
10403: else
10404: perror("uname() error");
1.179 brouard 10405: //#ifndef __INTEL_COMPILER
10406: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10407: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10408: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10409: #endif
1.169 brouard 10410: #endif
1.172 brouard 10411:
1.286 brouard 10412: // void main ()
1.172 brouard 10413: // {
1.169 brouard 10414: #if defined(_MSC_VER)
1.174 brouard 10415: if (IsWow64()){
1.191 brouard 10416: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10417: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10418: }
10419: else{
1.191 brouard 10420: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10421: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10422: }
1.172 brouard 10423: // printf("\nPress Enter to continue...");
10424: // getchar();
10425: // }
10426:
1.169 brouard 10427: #endif
10428:
1.167 brouard 10429:
1.219 brouard 10430: }
1.136 brouard 10431:
1.219 brouard 10432: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10433: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10434: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10435: /* double ftolpl = 1.e-10; */
1.180 brouard 10436: double age, agebase, agelim;
1.203 brouard 10437: double tot;
1.180 brouard 10438:
1.202 brouard 10439: strcpy(filerespl,"PL_");
10440: strcat(filerespl,fileresu);
10441: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10442: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10443: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10444: }
1.288 brouard 10445: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10446: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10447: pstamp(ficrespl);
1.288 brouard 10448: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10449: fprintf(ficrespl,"#Age ");
10450: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10451: fprintf(ficrespl,"\n");
1.180 brouard 10452:
1.219 brouard 10453: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10454:
1.219 brouard 10455: agebase=ageminpar;
10456: agelim=agemaxpar;
1.180 brouard 10457:
1.227 brouard 10458: /* i1=pow(2,ncoveff); */
1.234 brouard 10459: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10460: if (cptcovn < 1){i1=1;}
1.180 brouard 10461:
1.238 brouard 10462: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10463: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10464: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10465: continue;
1.235 brouard 10466:
1.238 brouard 10467: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10468: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10469: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10470: /* k=k+1; */
10471: /* to clean */
10472: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10473: fprintf(ficrespl,"#******");
10474: printf("#******");
10475: fprintf(ficlog,"#******");
10476: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10477: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10478: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10479: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10480: }
10481: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10482: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10483: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10484: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10485: }
10486: fprintf(ficrespl,"******\n");
10487: printf("******\n");
10488: fprintf(ficlog,"******\n");
10489: if(invalidvarcomb[k]){
10490: printf("\nCombination (%d) ignored because no case \n",k);
10491: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10492: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10493: continue;
10494: }
1.219 brouard 10495:
1.238 brouard 10496: fprintf(ficrespl,"#Age ");
10497: for(j=1;j<=cptcoveff;j++) {
10498: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10499: }
10500: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10501: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10502:
1.238 brouard 10503: for (age=agebase; age<=agelim; age++){
10504: /* for (age=agebase; age<=agebase; age++){ */
10505: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10506: fprintf(ficrespl,"%.0f ",age );
10507: for(j=1;j<=cptcoveff;j++)
10508: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10509: tot=0.;
10510: for(i=1; i<=nlstate;i++){
10511: tot += prlim[i][i];
10512: fprintf(ficrespl," %.5f", prlim[i][i]);
10513: }
10514: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10515: } /* Age */
10516: /* was end of cptcod */
10517: } /* cptcov */
10518: } /* nres */
1.219 brouard 10519: return 0;
1.180 brouard 10520: }
10521:
1.218 brouard 10522: 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 10523: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10524:
10525: /* Computes the back prevalence limit for any combination of covariate values
10526: * at any age between ageminpar and agemaxpar
10527: */
1.235 brouard 10528: int i, j, k, i1, nres=0 ;
1.217 brouard 10529: /* double ftolpl = 1.e-10; */
10530: double age, agebase, agelim;
10531: double tot;
1.218 brouard 10532: /* double ***mobaverage; */
10533: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10534:
10535: strcpy(fileresplb,"PLB_");
10536: strcat(fileresplb,fileresu);
10537: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10538: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10539: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10540: }
1.288 brouard 10541: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10542: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10543: pstamp(ficresplb);
1.288 brouard 10544: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10545: fprintf(ficresplb,"#Age ");
10546: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10547: fprintf(ficresplb,"\n");
10548:
1.218 brouard 10549:
10550: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10551:
10552: agebase=ageminpar;
10553: agelim=agemaxpar;
10554:
10555:
1.227 brouard 10556: i1=pow(2,cptcoveff);
1.218 brouard 10557: if (cptcovn < 1){i1=1;}
1.227 brouard 10558:
1.238 brouard 10559: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10560: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10561: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10562: continue;
10563: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10564: fprintf(ficresplb,"#******");
10565: printf("#******");
10566: fprintf(ficlog,"#******");
10567: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10568: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10569: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10570: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10571: }
10572: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10573: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10574: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10575: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10576: }
10577: fprintf(ficresplb,"******\n");
10578: printf("******\n");
10579: fprintf(ficlog,"******\n");
10580: if(invalidvarcomb[k]){
10581: printf("\nCombination (%d) ignored because no cases \n",k);
10582: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10583: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10584: continue;
10585: }
1.218 brouard 10586:
1.238 brouard 10587: fprintf(ficresplb,"#Age ");
10588: for(j=1;j<=cptcoveff;j++) {
10589: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10590: }
10591: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10592: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10593:
10594:
1.238 brouard 10595: for (age=agebase; age<=agelim; age++){
10596: /* for (age=agebase; age<=agebase; age++){ */
10597: if(mobilavproj > 0){
10598: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10599: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10600: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10601: }else if (mobilavproj == 0){
10602: 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);
10603: 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);
10604: exit(1);
10605: }else{
10606: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10607: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10608: /* printf("TOTOT\n"); */
10609: /* exit(1); */
1.238 brouard 10610: }
10611: fprintf(ficresplb,"%.0f ",age );
10612: for(j=1;j<=cptcoveff;j++)
10613: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10614: tot=0.;
10615: for(i=1; i<=nlstate;i++){
10616: tot += bprlim[i][i];
10617: fprintf(ficresplb," %.5f", bprlim[i][i]);
10618: }
10619: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10620: } /* Age */
10621: /* was end of cptcod */
1.255 brouard 10622: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10623: } /* end of any combination */
10624: } /* end of nres */
1.218 brouard 10625: /* hBijx(p, bage, fage); */
10626: /* fclose(ficrespijb); */
10627:
10628: return 0;
1.217 brouard 10629: }
1.218 brouard 10630:
1.180 brouard 10631: int hPijx(double *p, int bage, int fage){
10632: /*------------- h Pij x at various ages ------------*/
10633:
10634: int stepsize;
10635: int agelim;
10636: int hstepm;
10637: int nhstepm;
1.235 brouard 10638: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10639:
10640: double agedeb;
10641: double ***p3mat;
10642:
1.201 brouard 10643: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10644: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10645: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10646: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10647: }
10648: printf("Computing pij: result on file '%s' \n", filerespij);
10649: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10650:
10651: stepsize=(int) (stepm+YEARM-1)/YEARM;
10652: /*if (stepm<=24) stepsize=2;*/
10653:
10654: agelim=AGESUP;
10655: hstepm=stepsize*YEARM; /* Every year of age */
10656: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10657:
1.180 brouard 10658: /* hstepm=1; aff par mois*/
10659: pstamp(ficrespij);
10660: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10661: i1= pow(2,cptcoveff);
1.218 brouard 10662: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10663: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10664: /* k=k+1; */
1.235 brouard 10665: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10666: for(k=1; k<=i1;k++){
1.253 brouard 10667: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10668: continue;
1.183 brouard 10669: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10670: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10671: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10672: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10673: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10674: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10675: }
1.183 brouard 10676: fprintf(ficrespij,"******\n");
10677:
10678: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10679: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10680: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10681:
10682: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10683:
1.183 brouard 10684: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10685: oldm=oldms;savm=savms;
1.235 brouard 10686: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10687: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10688: for(i=1; i<=nlstate;i++)
10689: for(j=1; j<=nlstate+ndeath;j++)
10690: fprintf(ficrespij," %1d-%1d",i,j);
10691: fprintf(ficrespij,"\n");
10692: for (h=0; h<=nhstepm; h++){
10693: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10694: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10695: for(i=1; i<=nlstate;i++)
10696: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10697: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10698: fprintf(ficrespij,"\n");
10699: }
1.183 brouard 10700: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10701: fprintf(ficrespij,"\n");
10702: }
1.180 brouard 10703: /*}*/
10704: }
1.218 brouard 10705: return 0;
1.180 brouard 10706: }
1.218 brouard 10707:
10708: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10709: /*------------- h Bij x at various ages ------------*/
10710:
10711: int stepsize;
1.218 brouard 10712: /* int agelim; */
10713: int ageminl;
1.217 brouard 10714: int hstepm;
10715: int nhstepm;
1.238 brouard 10716: int h, i, i1, j, k, nres;
1.218 brouard 10717:
1.217 brouard 10718: double agedeb;
10719: double ***p3mat;
1.218 brouard 10720:
10721: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10722: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10723: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10724: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10725: }
10726: printf("Computing pij back: result on file '%s' \n", filerespijb);
10727: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10728:
10729: stepsize=(int) (stepm+YEARM-1)/YEARM;
10730: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10731:
1.218 brouard 10732: /* agelim=AGESUP; */
1.289 brouard 10733: ageminl=AGEINF; /* was 30 */
1.218 brouard 10734: hstepm=stepsize*YEARM; /* Every year of age */
10735: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10736:
10737: /* hstepm=1; aff par mois*/
10738: pstamp(ficrespijb);
1.255 brouard 10739: 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 10740: i1= pow(2,cptcoveff);
1.218 brouard 10741: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10742: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10743: /* k=k+1; */
1.238 brouard 10744: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10745: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10746: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10747: continue;
10748: fprintf(ficrespijb,"\n#****** ");
10749: for(j=1;j<=cptcoveff;j++)
10750: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10751: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10752: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10753: }
10754: fprintf(ficrespijb,"******\n");
1.264 brouard 10755: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10756: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10757: continue;
10758: }
10759:
10760: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10761: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10762: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10763: 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 */
10764: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10765:
10766: /* nhstepm=nhstepm*YEARM; aff par mois*/
10767:
1.266 brouard 10768: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10769: /* and memory limitations if stepm is small */
10770:
1.238 brouard 10771: /* oldm=oldms;savm=savms; */
10772: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10773: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10774: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10775: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10776: for(i=1; i<=nlstate;i++)
10777: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10778: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10779: fprintf(ficrespijb,"\n");
1.238 brouard 10780: for (h=0; h<=nhstepm; h++){
10781: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10782: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10783: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10784: for(i=1; i<=nlstate;i++)
10785: for(j=1; j<=nlstate+ndeath;j++)
10786: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10787: fprintf(ficrespijb,"\n");
10788: }
10789: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10790: fprintf(ficrespijb,"\n");
10791: } /* end age deb */
10792: } /* end combination */
10793: } /* end nres */
1.218 brouard 10794: return 0;
10795: } /* hBijx */
1.217 brouard 10796:
1.180 brouard 10797:
1.136 brouard 10798: /***********************************************/
10799: /**************** Main Program *****************/
10800: /***********************************************/
10801:
10802: int main(int argc, char *argv[])
10803: {
10804: #ifdef GSL
10805: const gsl_multimin_fminimizer_type *T;
10806: size_t iteri = 0, it;
10807: int rval = GSL_CONTINUE;
10808: int status = GSL_SUCCESS;
10809: double ssval;
10810: #endif
10811: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10812: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10813: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10814: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10815: int jj, ll, li, lj, lk;
1.136 brouard 10816: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10817: int num_filled;
1.136 brouard 10818: int itimes;
10819: int NDIM=2;
10820: int vpopbased=0;
1.235 brouard 10821: int nres=0;
1.258 brouard 10822: int endishere=0;
1.277 brouard 10823: int noffset=0;
1.274 brouard 10824: int ncurrv=0; /* Temporary variable */
10825:
1.164 brouard 10826: char ca[32], cb[32];
1.136 brouard 10827: /* FILE *fichtm; *//* Html File */
10828: /* FILE *ficgp;*/ /*Gnuplot File */
10829: struct stat info;
1.191 brouard 10830: double agedeb=0.;
1.194 brouard 10831:
10832: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10833: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10834:
1.165 brouard 10835: double fret;
1.191 brouard 10836: double dum=0.; /* Dummy variable */
1.136 brouard 10837: double ***p3mat;
1.218 brouard 10838: /* double ***mobaverage; */
1.164 brouard 10839:
10840: char line[MAXLINE];
1.197 brouard 10841: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10842:
1.234 brouard 10843: char modeltemp[MAXLINE];
1.230 brouard 10844: char resultline[MAXLINE];
10845:
1.136 brouard 10846: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10847: char *tok, *val; /* pathtot */
1.290 brouard 10848: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10849: int c, h , cpt, c2;
1.191 brouard 10850: int jl=0;
10851: int i1, j1, jk, stepsize=0;
1.194 brouard 10852: int count=0;
10853:
1.164 brouard 10854: int *tab;
1.136 brouard 10855: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10856: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10857: /* double anprojf, mprojf, jprojf; */
10858: /* double jintmean,mintmean,aintmean; */
10859: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10860: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10861: double yrfproj= 10.0; /* Number of years of forward projections */
10862: double yrbproj= 10.0; /* Number of years of backward projections */
10863: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 10864: int mobilav=0,popforecast=0;
1.191 brouard 10865: int hstepm=0, nhstepm=0;
1.136 brouard 10866: int agemortsup;
10867: float sumlpop=0.;
10868: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10869: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10870:
1.191 brouard 10871: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10872: double ftolpl=FTOL;
10873: double **prlim;
1.217 brouard 10874: double **bprlim;
1.136 brouard 10875: double ***param; /* Matrix of parameters */
1.251 brouard 10876: double ***paramstart; /* Matrix of starting parameter values */
10877: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10878: double **matcov; /* Matrix of covariance */
1.203 brouard 10879: double **hess; /* Hessian matrix */
1.136 brouard 10880: double ***delti3; /* Scale */
10881: double *delti; /* Scale */
10882: double ***eij, ***vareij;
10883: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10884:
1.136 brouard 10885: double *epj, vepp;
1.164 brouard 10886:
1.273 brouard 10887: double dateprev1, dateprev2;
1.296 brouard 10888: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
10889: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
10890:
1.217 brouard 10891:
1.136 brouard 10892: double **ximort;
1.145 brouard 10893: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10894: int *dcwave;
10895:
1.164 brouard 10896: char z[1]="c";
1.136 brouard 10897:
10898: /*char *strt;*/
10899: char strtend[80];
1.126 brouard 10900:
1.164 brouard 10901:
1.126 brouard 10902: /* setlocale (LC_ALL, ""); */
10903: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10904: /* textdomain (PACKAGE); */
10905: /* setlocale (LC_CTYPE, ""); */
10906: /* setlocale (LC_MESSAGES, ""); */
10907:
10908: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10909: rstart_time = time(NULL);
10910: /* (void) gettimeofday(&start_time,&tzp);*/
10911: start_time = *localtime(&rstart_time);
1.126 brouard 10912: curr_time=start_time;
1.157 brouard 10913: /*tml = *localtime(&start_time.tm_sec);*/
10914: /* strcpy(strstart,asctime(&tml)); */
10915: strcpy(strstart,asctime(&start_time));
1.126 brouard 10916:
10917: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10918: /* tp.tm_sec = tp.tm_sec +86400; */
10919: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10920: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10921: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10922: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10923: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10924: /* strt=asctime(&tmg); */
10925: /* printf("Time(after) =%s",strstart); */
10926: /* (void) time (&time_value);
10927: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10928: * tm = *localtime(&time_value);
10929: * strstart=asctime(&tm);
10930: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10931: */
10932:
10933: nberr=0; /* Number of errors and warnings */
10934: nbwarn=0;
1.184 brouard 10935: #ifdef WIN32
10936: _getcwd(pathcd, size);
10937: #else
1.126 brouard 10938: getcwd(pathcd, size);
1.184 brouard 10939: #endif
1.191 brouard 10940: syscompilerinfo(0);
1.196 brouard 10941: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10942: if(argc <=1){
10943: printf("\nEnter the parameter file name: ");
1.205 brouard 10944: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10945: printf("ERROR Empty parameter file name\n");
10946: goto end;
10947: }
1.126 brouard 10948: i=strlen(pathr);
10949: if(pathr[i-1]=='\n')
10950: pathr[i-1]='\0';
1.156 brouard 10951: i=strlen(pathr);
1.205 brouard 10952: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10953: pathr[i-1]='\0';
1.205 brouard 10954: }
10955: i=strlen(pathr);
10956: if( i==0 ){
10957: printf("ERROR Empty parameter file name\n");
10958: goto end;
10959: }
10960: for (tok = pathr; tok != NULL; ){
1.126 brouard 10961: printf("Pathr |%s|\n",pathr);
10962: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10963: printf("val= |%s| pathr=%s\n",val,pathr);
10964: strcpy (pathtot, val);
10965: if(pathr[0] == '\0') break; /* Dirty */
10966: }
10967: }
1.281 brouard 10968: else if (argc<=2){
10969: strcpy(pathtot,argv[1]);
10970: }
1.126 brouard 10971: else{
10972: strcpy(pathtot,argv[1]);
1.281 brouard 10973: strcpy(z,argv[2]);
10974: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 10975: }
10976: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10977: /*cygwin_split_path(pathtot,path,optionfile);
10978: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10979: /* cutv(path,optionfile,pathtot,'\\');*/
10980:
10981: /* Split argv[0], imach program to get pathimach */
10982: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10983: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10984: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10985: /* strcpy(pathimach,argv[0]); */
10986: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10987: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10988: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10989: #ifdef WIN32
10990: _chdir(path); /* Can be a relative path */
10991: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10992: #else
1.126 brouard 10993: chdir(path); /* Can be a relative path */
1.184 brouard 10994: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10995: #endif
10996: printf("Current directory %s!\n",pathcd);
1.126 brouard 10997: strcpy(command,"mkdir ");
10998: strcat(command,optionfilefiname);
10999: if((outcmd=system(command)) != 0){
1.169 brouard 11000: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11001: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11002: /* fclose(ficlog); */
11003: /* exit(1); */
11004: }
11005: /* if((imk=mkdir(optionfilefiname))<0){ */
11006: /* perror("mkdir"); */
11007: /* } */
11008:
11009: /*-------- arguments in the command line --------*/
11010:
1.186 brouard 11011: /* Main Log file */
1.126 brouard 11012: strcat(filelog, optionfilefiname);
11013: strcat(filelog,".log"); /* */
11014: if((ficlog=fopen(filelog,"w"))==NULL) {
11015: printf("Problem with logfile %s\n",filelog);
11016: goto end;
11017: }
11018: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11019: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11020: fprintf(ficlog,"\nEnter the parameter file name: \n");
11021: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11022: path=%s \n\
11023: optionfile=%s\n\
11024: optionfilext=%s\n\
1.156 brouard 11025: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11026:
1.197 brouard 11027: syscompilerinfo(1);
1.167 brouard 11028:
1.126 brouard 11029: printf("Local time (at start):%s",strstart);
11030: fprintf(ficlog,"Local time (at start): %s",strstart);
11031: fflush(ficlog);
11032: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11033: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11034:
11035: /* */
11036: strcpy(fileres,"r");
11037: strcat(fileres, optionfilefiname);
1.201 brouard 11038: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11039: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11040: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11041:
1.186 brouard 11042: /* Main ---------arguments file --------*/
1.126 brouard 11043:
11044: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11045: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11046: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11047: fflush(ficlog);
1.149 brouard 11048: /* goto end; */
11049: exit(70);
1.126 brouard 11050: }
11051:
11052: strcpy(filereso,"o");
1.201 brouard 11053: strcat(filereso,fileresu);
1.126 brouard 11054: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11055: printf("Problem with Output resultfile: %s\n", filereso);
11056: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11057: fflush(ficlog);
11058: goto end;
11059: }
1.278 brouard 11060: /*-------- Rewriting parameter file ----------*/
11061: strcpy(rfileres,"r"); /* "Rparameterfile */
11062: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11063: strcat(rfileres,"."); /* */
11064: strcat(rfileres,optionfilext); /* Other files have txt extension */
11065: if((ficres =fopen(rfileres,"w"))==NULL) {
11066: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11067: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11068: fflush(ficlog);
11069: goto end;
11070: }
11071: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11072:
1.278 brouard 11073:
1.126 brouard 11074: /* Reads comments: lines beginning with '#' */
11075: numlinepar=0;
1.277 brouard 11076: /* Is it a BOM UTF-8 Windows file? */
11077: /* First parameter line */
1.197 brouard 11078: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11079: noffset=0;
11080: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11081: {
11082: noffset=noffset+3;
11083: printf("# File is an UTF8 Bom.\n"); // 0xBF
11084: }
11085: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
11086: {
11087: noffset=noffset+2;
11088: printf("# File is an UTF16BE BOM file\n");
11089: }
11090: else if( line[0] == 0 && line[1] == 0)
11091: {
11092: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11093: noffset=noffset+4;
11094: printf("# File is an UTF16BE BOM file\n");
11095: }
11096: } else{
11097: ;/*printf(" Not a BOM file\n");*/
11098: }
11099:
1.197 brouard 11100: /* If line starts with a # it is a comment */
1.277 brouard 11101: if (line[noffset] == '#') {
1.197 brouard 11102: numlinepar++;
11103: fputs(line,stdout);
11104: fputs(line,ficparo);
1.278 brouard 11105: fputs(line,ficres);
1.197 brouard 11106: fputs(line,ficlog);
11107: continue;
11108: }else
11109: break;
11110: }
11111: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11112: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11113: if (num_filled != 5) {
11114: printf("Should be 5 parameters\n");
1.283 brouard 11115: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11116: }
1.126 brouard 11117: numlinepar++;
1.197 brouard 11118: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11119: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11120: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11121: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11122: }
11123: /* Second parameter line */
11124: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11125: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11126: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11127: if (line[0] == '#') {
11128: numlinepar++;
1.283 brouard 11129: printf("%s",line);
11130: fprintf(ficres,"%s",line);
11131: fprintf(ficparo,"%s",line);
11132: fprintf(ficlog,"%s",line);
1.197 brouard 11133: continue;
11134: }else
11135: break;
11136: }
1.223 brouard 11137: 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", \
11138: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11139: if (num_filled != 11) {
11140: 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 11141: printf("but line=%s\n",line);
1.283 brouard 11142: 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");
11143: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11144: }
1.286 brouard 11145: if( lastpass > maxwav){
11146: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11147: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11148: fflush(ficlog);
11149: goto end;
11150: }
11151: 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 11152: 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 11153: 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 11154: 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 11155: }
1.203 brouard 11156: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11157: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11158: /* Third parameter line */
11159: while(fgets(line, MAXLINE, ficpar)) {
11160: /* If line starts with a # it is a comment */
11161: if (line[0] == '#') {
11162: numlinepar++;
1.283 brouard 11163: printf("%s",line);
11164: fprintf(ficres,"%s",line);
11165: fprintf(ficparo,"%s",line);
11166: fprintf(ficlog,"%s",line);
1.197 brouard 11167: continue;
11168: }else
11169: break;
11170: }
1.201 brouard 11171: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11172: if (num_filled != 1){
11173: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
11174: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 11175: model[0]='\0';
11176: goto end;
11177: }
11178: else{
11179: if (model[0]=='+'){
11180: for(i=1; i<=strlen(model);i++)
11181: modeltemp[i-1]=model[i];
1.201 brouard 11182: strcpy(model,modeltemp);
1.197 brouard 11183: }
11184: }
1.199 brouard 11185: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11186: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11187: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11188: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11189: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11190: }
11191: /* 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); */
11192: /* numlinepar=numlinepar+3; /\* In general *\/ */
11193: /* 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 11194: /* 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); */
11195: /* 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 11196: fflush(ficlog);
1.190 brouard 11197: /* if(model[0]=='#'|| model[0]== '\0'){ */
11198: if(model[0]=='#'){
1.279 brouard 11199: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11200: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11201: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11202: if(mle != -1){
1.279 brouard 11203: 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 11204: exit(1);
11205: }
11206: }
1.126 brouard 11207: while((c=getc(ficpar))=='#' && c!= EOF){
11208: ungetc(c,ficpar);
11209: fgets(line, MAXLINE, ficpar);
11210: numlinepar++;
1.195 brouard 11211: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11212: z[0]=line[1];
11213: }
11214: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11215: fputs(line, stdout);
11216: //puts(line);
1.126 brouard 11217: fputs(line,ficparo);
11218: fputs(line,ficlog);
11219: }
11220: ungetc(c,ficpar);
11221:
11222:
1.290 brouard 11223: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11224: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11225: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11226: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11227: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11228: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11229: v1+v2*age+v2*v3 makes cptcovn = 3
11230: */
11231: if (strlen(model)>1)
1.187 brouard 11232: 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 11233: else
1.187 brouard 11234: ncovmodel=2; /* Constant and age */
1.133 brouard 11235: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11236: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11237: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11238: 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);
11239: 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);
11240: fflush(stdout);
11241: fclose (ficlog);
11242: goto end;
11243: }
1.126 brouard 11244: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11245: delti=delti3[1][1];
11246: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11247: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11248: /* We could also provide initial parameters values giving by simple logistic regression
11249: * only one way, that is without matrix product. We will have nlstate maximizations */
11250: /* for(i=1;i<nlstate;i++){ */
11251: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11252: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11253: /* } */
1.126 brouard 11254: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11255: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11256: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11257: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11258: fclose (ficparo);
11259: fclose (ficlog);
11260: goto end;
11261: exit(0);
1.220 brouard 11262: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11263: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11264: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11265: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11266: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11267: matcov=matrix(1,npar,1,npar);
1.203 brouard 11268: hess=matrix(1,npar,1,npar);
1.220 brouard 11269: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11270: /* Read guessed parameters */
1.126 brouard 11271: /* Reads comments: lines beginning with '#' */
11272: while((c=getc(ficpar))=='#' && c!= EOF){
11273: ungetc(c,ficpar);
11274: fgets(line, MAXLINE, ficpar);
11275: numlinepar++;
1.141 brouard 11276: fputs(line,stdout);
1.126 brouard 11277: fputs(line,ficparo);
11278: fputs(line,ficlog);
11279: }
11280: ungetc(c,ficpar);
11281:
11282: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11283: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11284: for(i=1; i <=nlstate; i++){
1.234 brouard 11285: j=0;
1.126 brouard 11286: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11287: if(jj==i) continue;
11288: j++;
1.292 brouard 11289: while((c=getc(ficpar))=='#' && c!= EOF){
11290: ungetc(c,ficpar);
11291: fgets(line, MAXLINE, ficpar);
11292: numlinepar++;
11293: fputs(line,stdout);
11294: fputs(line,ficparo);
11295: fputs(line,ficlog);
11296: }
11297: ungetc(c,ficpar);
1.234 brouard 11298: fscanf(ficpar,"%1d%1d",&i1,&j1);
11299: if ((i1 != i) || (j1 != jj)){
11300: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11301: It might be a problem of design; if ncovcol and the model are correct\n \
11302: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11303: exit(1);
11304: }
11305: fprintf(ficparo,"%1d%1d",i1,j1);
11306: if(mle==1)
11307: printf("%1d%1d",i,jj);
11308: fprintf(ficlog,"%1d%1d",i,jj);
11309: for(k=1; k<=ncovmodel;k++){
11310: fscanf(ficpar," %lf",¶m[i][j][k]);
11311: if(mle==1){
11312: printf(" %lf",param[i][j][k]);
11313: fprintf(ficlog," %lf",param[i][j][k]);
11314: }
11315: else
11316: fprintf(ficlog," %lf",param[i][j][k]);
11317: fprintf(ficparo," %lf",param[i][j][k]);
11318: }
11319: fscanf(ficpar,"\n");
11320: numlinepar++;
11321: if(mle==1)
11322: printf("\n");
11323: fprintf(ficlog,"\n");
11324: fprintf(ficparo,"\n");
1.126 brouard 11325: }
11326: }
11327: fflush(ficlog);
1.234 brouard 11328:
1.251 brouard 11329: /* Reads parameters values */
1.126 brouard 11330: p=param[1][1];
1.251 brouard 11331: pstart=paramstart[1][1];
1.126 brouard 11332:
11333: /* Reads comments: lines beginning with '#' */
11334: while((c=getc(ficpar))=='#' && c!= EOF){
11335: ungetc(c,ficpar);
11336: fgets(line, MAXLINE, ficpar);
11337: numlinepar++;
1.141 brouard 11338: fputs(line,stdout);
1.126 brouard 11339: fputs(line,ficparo);
11340: fputs(line,ficlog);
11341: }
11342: ungetc(c,ficpar);
11343:
11344: for(i=1; i <=nlstate; i++){
11345: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11346: fscanf(ficpar,"%1d%1d",&i1,&j1);
11347: if ( (i1-i) * (j1-j) != 0){
11348: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11349: exit(1);
11350: }
11351: printf("%1d%1d",i,j);
11352: fprintf(ficparo,"%1d%1d",i1,j1);
11353: fprintf(ficlog,"%1d%1d",i1,j1);
11354: for(k=1; k<=ncovmodel;k++){
11355: fscanf(ficpar,"%le",&delti3[i][j][k]);
11356: printf(" %le",delti3[i][j][k]);
11357: fprintf(ficparo," %le",delti3[i][j][k]);
11358: fprintf(ficlog," %le",delti3[i][j][k]);
11359: }
11360: fscanf(ficpar,"\n");
11361: numlinepar++;
11362: printf("\n");
11363: fprintf(ficparo,"\n");
11364: fprintf(ficlog,"\n");
1.126 brouard 11365: }
11366: }
11367: fflush(ficlog);
1.234 brouard 11368:
1.145 brouard 11369: /* Reads covariance matrix */
1.126 brouard 11370: delti=delti3[1][1];
1.220 brouard 11371:
11372:
1.126 brouard 11373: /* 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 11374:
1.126 brouard 11375: /* Reads comments: lines beginning with '#' */
11376: while((c=getc(ficpar))=='#' && c!= EOF){
11377: ungetc(c,ficpar);
11378: fgets(line, MAXLINE, ficpar);
11379: numlinepar++;
1.141 brouard 11380: fputs(line,stdout);
1.126 brouard 11381: fputs(line,ficparo);
11382: fputs(line,ficlog);
11383: }
11384: ungetc(c,ficpar);
1.220 brouard 11385:
1.126 brouard 11386: matcov=matrix(1,npar,1,npar);
1.203 brouard 11387: hess=matrix(1,npar,1,npar);
1.131 brouard 11388: for(i=1; i <=npar; i++)
11389: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11390:
1.194 brouard 11391: /* Scans npar lines */
1.126 brouard 11392: for(i=1; i <=npar; i++){
1.226 brouard 11393: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11394: if(count != 3){
1.226 brouard 11395: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11396: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11397: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11398: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11399: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11400: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11401: exit(1);
1.220 brouard 11402: }else{
1.226 brouard 11403: if(mle==1)
11404: printf("%1d%1d%d",i1,j1,jk);
11405: }
11406: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11407: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11408: for(j=1; j <=i; j++){
1.226 brouard 11409: fscanf(ficpar," %le",&matcov[i][j]);
11410: if(mle==1){
11411: printf(" %.5le",matcov[i][j]);
11412: }
11413: fprintf(ficlog," %.5le",matcov[i][j]);
11414: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11415: }
11416: fscanf(ficpar,"\n");
11417: numlinepar++;
11418: if(mle==1)
1.220 brouard 11419: printf("\n");
1.126 brouard 11420: fprintf(ficlog,"\n");
11421: fprintf(ficparo,"\n");
11422: }
1.194 brouard 11423: /* End of read covariance matrix npar lines */
1.126 brouard 11424: for(i=1; i <=npar; i++)
11425: for(j=i+1;j<=npar;j++)
1.226 brouard 11426: matcov[i][j]=matcov[j][i];
1.126 brouard 11427:
11428: if(mle==1)
11429: printf("\n");
11430: fprintf(ficlog,"\n");
11431:
11432: fflush(ficlog);
11433:
11434: } /* End of mle != -3 */
1.218 brouard 11435:
1.186 brouard 11436: /* Main data
11437: */
1.290 brouard 11438: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11439: /* num=lvector(1,n); */
11440: /* moisnais=vector(1,n); */
11441: /* annais=vector(1,n); */
11442: /* moisdc=vector(1,n); */
11443: /* andc=vector(1,n); */
11444: /* weight=vector(1,n); */
11445: /* agedc=vector(1,n); */
11446: /* cod=ivector(1,n); */
11447: /* for(i=1;i<=n;i++){ */
11448: num=lvector(firstobs,lastobs);
11449: moisnais=vector(firstobs,lastobs);
11450: annais=vector(firstobs,lastobs);
11451: moisdc=vector(firstobs,lastobs);
11452: andc=vector(firstobs,lastobs);
11453: weight=vector(firstobs,lastobs);
11454: agedc=vector(firstobs,lastobs);
11455: cod=ivector(firstobs,lastobs);
11456: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11457: num[i]=0;
11458: moisnais[i]=0;
11459: annais[i]=0;
11460: moisdc[i]=0;
11461: andc[i]=0;
11462: agedc[i]=0;
11463: cod[i]=0;
11464: weight[i]=1.0; /* Equal weights, 1 by default */
11465: }
1.290 brouard 11466: mint=matrix(1,maxwav,firstobs,lastobs);
11467: anint=matrix(1,maxwav,firstobs,lastobs);
11468: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11469: tab=ivector(1,NCOVMAX);
1.144 brouard 11470: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11471: 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 11472:
1.136 brouard 11473: /* Reads data from file datafile */
11474: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11475: goto end;
11476:
11477: /* Calculation of the number of parameters from char model */
1.234 brouard 11478: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11479: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11480: k=3 V4 Tvar[k=3]= 4 (from V4)
11481: k=2 V1 Tvar[k=2]= 1 (from V1)
11482: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11483: */
11484:
11485: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11486: TvarsDind=ivector(1,NCOVMAX); /* */
11487: TvarsD=ivector(1,NCOVMAX); /* */
11488: TvarsQind=ivector(1,NCOVMAX); /* */
11489: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11490: TvarF=ivector(1,NCOVMAX); /* */
11491: TvarFind=ivector(1,NCOVMAX); /* */
11492: TvarV=ivector(1,NCOVMAX); /* */
11493: TvarVind=ivector(1,NCOVMAX); /* */
11494: TvarA=ivector(1,NCOVMAX); /* */
11495: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11496: TvarFD=ivector(1,NCOVMAX); /* */
11497: TvarFDind=ivector(1,NCOVMAX); /* */
11498: TvarFQ=ivector(1,NCOVMAX); /* */
11499: TvarFQind=ivector(1,NCOVMAX); /* */
11500: TvarVD=ivector(1,NCOVMAX); /* */
11501: TvarVDind=ivector(1,NCOVMAX); /* */
11502: TvarVQ=ivector(1,NCOVMAX); /* */
11503: TvarVQind=ivector(1,NCOVMAX); /* */
11504:
1.230 brouard 11505: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11506: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11507: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11508: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11509: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11510: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11511: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11512: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11513: */
11514: /* For model-covariate k tells which data-covariate to use but
11515: because this model-covariate is a construction we invent a new column
11516: ncovcol + k1
11517: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11518: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11519: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11520: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11521: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11522: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11523: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11524: */
1.145 brouard 11525: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11526: 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 11527: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11528: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11529: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11530: 4 covariates (3 plus signs)
11531: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11532: */
1.230 brouard 11533: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11534: * individual dummy, fixed or varying:
11535: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11536: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11537: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11538: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11539: * Tmodelind[1]@9={9,0,3,2,}*/
11540: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11541: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11542: * individual quantitative, fixed or varying:
11543: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11544: * 3, 1, 0, 0, 0, 0, 0, 0},
11545: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11546: /* Main decodemodel */
11547:
1.187 brouard 11548:
1.223 brouard 11549: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11550: goto end;
11551:
1.137 brouard 11552: if((double)(lastobs-imx)/(double)imx > 1.10){
11553: nbwarn++;
11554: 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);
11555: 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);
11556: }
1.136 brouard 11557: /* if(mle==1){*/
1.137 brouard 11558: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11559: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11560: }
11561:
11562: /*-calculation of age at interview from date of interview and age at death -*/
11563: agev=matrix(1,maxwav,1,imx);
11564:
11565: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11566: goto end;
11567:
1.126 brouard 11568:
1.136 brouard 11569: agegomp=(int)agemin;
1.290 brouard 11570: free_vector(moisnais,firstobs,lastobs);
11571: free_vector(annais,firstobs,lastobs);
1.126 brouard 11572: /* free_matrix(mint,1,maxwav,1,n);
11573: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11574: /* free_vector(moisdc,1,n); */
11575: /* free_vector(andc,1,n); */
1.145 brouard 11576: /* */
11577:
1.126 brouard 11578: wav=ivector(1,imx);
1.214 brouard 11579: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11580: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11581: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11582: 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.*/
11583: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11584: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11585:
11586: /* Concatenates waves */
1.214 brouard 11587: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11588: Death is a valid wave (if date is known).
11589: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11590: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11591: and mw[mi+1][i]. dh depends on stepm.
11592: */
11593:
1.126 brouard 11594: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11595: /* Concatenates waves */
1.145 brouard 11596:
1.290 brouard 11597: free_vector(moisdc,firstobs,lastobs);
11598: free_vector(andc,firstobs,lastobs);
1.215 brouard 11599:
1.126 brouard 11600: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11601: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11602: ncodemax[1]=1;
1.145 brouard 11603: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11604: cptcoveff=0;
1.220 brouard 11605: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11606: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11607: }
11608:
11609: ncovcombmax=pow(2,cptcoveff);
11610: invalidvarcomb=ivector(1, ncovcombmax);
11611: for(i=1;i<ncovcombmax;i++)
11612: invalidvarcomb[i]=0;
11613:
1.211 brouard 11614: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11615: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11616: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11617:
1.200 brouard 11618: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11619: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11620: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11621: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11622: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11623: * (currently 0 or 1) in the data.
11624: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11625: * corresponding modality (h,j).
11626: */
11627:
1.145 brouard 11628: h=0;
11629: /*if (cptcovn > 0) */
1.126 brouard 11630: m=pow(2,cptcoveff);
11631:
1.144 brouard 11632: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11633: * For k=4 covariates, h goes from 1 to m=2**k
11634: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11635: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11636: * h\k 1 2 3 4
1.143 brouard 11637: *______________________________
11638: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11639: * 2 2 1 1 1
11640: * 3 i=2 1 2 1 1
11641: * 4 2 2 1 1
11642: * 5 i=3 1 i=2 1 2 1
11643: * 6 2 1 2 1
11644: * 7 i=4 1 2 2 1
11645: * 8 2 2 2 1
1.197 brouard 11646: * 9 i=5 1 i=3 1 i=2 1 2
11647: * 10 2 1 1 2
11648: * 11 i=6 1 2 1 2
11649: * 12 2 2 1 2
11650: * 13 i=7 1 i=4 1 2 2
11651: * 14 2 1 2 2
11652: * 15 i=8 1 2 2 2
11653: * 16 2 2 2 2
1.143 brouard 11654: */
1.212 brouard 11655: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11656: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11657: * and the value of each covariate?
11658: * V1=1, V2=1, V3=2, V4=1 ?
11659: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11660: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11661: * In order to get the real value in the data, we use nbcode
11662: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11663: * We are keeping this crazy system in order to be able (in the future?)
11664: * to have more than 2 values (0 or 1) for a covariate.
11665: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11666: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11667: * bbbbbbbb
11668: * 76543210
11669: * h-1 00000101 (6-1=5)
1.219 brouard 11670: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11671: * &
11672: * 1 00000001 (1)
1.219 brouard 11673: * 00000000 = 1 & ((h-1) >> (k-1))
11674: * +1= 00000001 =1
1.211 brouard 11675: *
11676: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11677: * h' 1101 =2^3+2^2+0x2^1+2^0
11678: * >>k' 11
11679: * & 00000001
11680: * = 00000001
11681: * +1 = 00000010=2 = codtabm(14,3)
11682: * Reverse h=6 and m=16?
11683: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11684: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11685: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11686: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11687: * V3=decodtabm(14,3,2**4)=2
11688: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11689: *(h-1) >> (j-1) 0011 =13 >> 2
11690: * &1 000000001
11691: * = 000000001
11692: * +1= 000000010 =2
11693: * 2211
11694: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11695: * V3=2
1.220 brouard 11696: * codtabm and decodtabm are identical
1.211 brouard 11697: */
11698:
1.145 brouard 11699:
11700: free_ivector(Ndum,-1,NCOVMAX);
11701:
11702:
1.126 brouard 11703:
1.186 brouard 11704: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11705: strcpy(optionfilegnuplot,optionfilefiname);
11706: if(mle==-3)
1.201 brouard 11707: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11708: strcat(optionfilegnuplot,".gp");
11709:
11710: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11711: printf("Problem with file %s",optionfilegnuplot);
11712: }
11713: else{
1.204 brouard 11714: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11715: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11716: //fprintf(ficgp,"set missing 'NaNq'\n");
11717: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11718: }
11719: /* fclose(ficgp);*/
1.186 brouard 11720:
11721:
11722: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11723:
11724: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11725: if(mle==-3)
1.201 brouard 11726: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11727: strcat(optionfilehtm,".htm");
11728: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11729: printf("Problem with %s \n",optionfilehtm);
11730: exit(0);
1.126 brouard 11731: }
11732:
11733: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11734: strcat(optionfilehtmcov,"-cov.htm");
11735: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11736: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11737: }
11738: else{
11739: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11740: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11741: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11742: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11743: }
11744:
1.213 brouard 11745: 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å¦è¡“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 11746: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11747: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11748: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11749: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11750: \n\
11751: <hr size=\"2\" color=\"#EC5E5E\">\
11752: <ul><li><h4>Parameter files</h4>\n\
11753: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11754: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11755: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11756: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11757: - Date and time at start: %s</ul>\n",\
11758: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11759: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11760: fileres,fileres,\
11761: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11762: fflush(fichtm);
11763:
11764: strcpy(pathr,path);
11765: strcat(pathr,optionfilefiname);
1.184 brouard 11766: #ifdef WIN32
11767: _chdir(optionfilefiname); /* Move to directory named optionfile */
11768: #else
1.126 brouard 11769: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11770: #endif
11771:
1.126 brouard 11772:
1.220 brouard 11773: /* Calculates basic frequencies. Computes observed prevalence at single age
11774: and for any valid combination of covariates
1.126 brouard 11775: and prints on file fileres'p'. */
1.251 brouard 11776: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11777: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11778:
11779: fprintf(fichtm,"\n");
1.286 brouard 11780: 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 11781: ftol, stepm);
11782: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11783: ncurrv=1;
11784: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11785: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11786: ncurrv=i;
11787: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11788: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11789: ncurrv=i;
11790: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11791: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11792: ncurrv=i;
11793: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11794: 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", \
11795: nlstate, ndeath, maxwav, mle, weightopt);
11796:
11797: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11798: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11799:
11800:
11801: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11802: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11803: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11804: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11805: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11806: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11807: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11808: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11809: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11810:
1.126 brouard 11811: /* For Powell, parameters are in a vector p[] starting at p[1]
11812: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11813: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11814:
11815: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11816: /* For mortality only */
1.126 brouard 11817: if (mle==-3){
1.136 brouard 11818: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11819: for(i=1;i<=NDIM;i++)
11820: for(j=1;j<=NDIM;j++)
11821: ximort[i][j]=0.;
1.186 brouard 11822: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11823: cens=ivector(firstobs,lastobs);
11824: ageexmed=vector(firstobs,lastobs);
11825: agecens=vector(firstobs,lastobs);
11826: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11827:
1.126 brouard 11828: for (i=1; i<=imx; i++){
11829: dcwave[i]=-1;
11830: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11831: if (s[m][i]>nlstate) {
11832: dcwave[i]=m;
11833: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11834: break;
11835: }
1.126 brouard 11836: }
1.226 brouard 11837:
1.126 brouard 11838: for (i=1; i<=imx; i++) {
11839: if (wav[i]>0){
1.226 brouard 11840: ageexmed[i]=agev[mw[1][i]][i];
11841: j=wav[i];
11842: agecens[i]=1.;
11843:
11844: if (ageexmed[i]> 1 && wav[i] > 0){
11845: agecens[i]=agev[mw[j][i]][i];
11846: cens[i]= 1;
11847: }else if (ageexmed[i]< 1)
11848: cens[i]= -1;
11849: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11850: cens[i]=0 ;
1.126 brouard 11851: }
11852: else cens[i]=-1;
11853: }
11854:
11855: for (i=1;i<=NDIM;i++) {
11856: for (j=1;j<=NDIM;j++)
1.226 brouard 11857: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11858: }
11859:
1.145 brouard 11860: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11861: /*printf("%lf %lf", p[1], p[2]);*/
11862:
11863:
1.136 brouard 11864: #ifdef GSL
11865: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11866: #else
1.126 brouard 11867: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11868: #endif
1.201 brouard 11869: strcpy(filerespow,"POW-MORT_");
11870: strcat(filerespow,fileresu);
1.126 brouard 11871: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11872: printf("Problem with resultfile: %s\n", filerespow);
11873: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11874: }
1.136 brouard 11875: #ifdef GSL
11876: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11877: #else
1.126 brouard 11878: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11879: #endif
1.126 brouard 11880: /* for (i=1;i<=nlstate;i++)
11881: for(j=1;j<=nlstate+ndeath;j++)
11882: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11883: */
11884: fprintf(ficrespow,"\n");
1.136 brouard 11885: #ifdef GSL
11886: /* gsl starts here */
11887: T = gsl_multimin_fminimizer_nmsimplex;
11888: gsl_multimin_fminimizer *sfm = NULL;
11889: gsl_vector *ss, *x;
11890: gsl_multimin_function minex_func;
11891:
11892: /* Initial vertex size vector */
11893: ss = gsl_vector_alloc (NDIM);
11894:
11895: if (ss == NULL){
11896: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11897: }
11898: /* Set all step sizes to 1 */
11899: gsl_vector_set_all (ss, 0.001);
11900:
11901: /* Starting point */
1.126 brouard 11902:
1.136 brouard 11903: x = gsl_vector_alloc (NDIM);
11904:
11905: if (x == NULL){
11906: gsl_vector_free(ss);
11907: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11908: }
11909:
11910: /* Initialize method and iterate */
11911: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11912: /* gsl_vector_set(x, 0, 0.0268); */
11913: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11914: gsl_vector_set(x, 0, p[1]);
11915: gsl_vector_set(x, 1, p[2]);
11916:
11917: minex_func.f = &gompertz_f;
11918: minex_func.n = NDIM;
11919: minex_func.params = (void *)&p; /* ??? */
11920:
11921: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11922: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11923:
11924: printf("Iterations beginning .....\n\n");
11925: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11926:
11927: iteri=0;
11928: while (rval == GSL_CONTINUE){
11929: iteri++;
11930: status = gsl_multimin_fminimizer_iterate(sfm);
11931:
11932: if (status) printf("error: %s\n", gsl_strerror (status));
11933: fflush(0);
11934:
11935: if (status)
11936: break;
11937:
11938: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11939: ssval = gsl_multimin_fminimizer_size (sfm);
11940:
11941: if (rval == GSL_SUCCESS)
11942: printf ("converged to a local maximum at\n");
11943:
11944: printf("%5d ", iteri);
11945: for (it = 0; it < NDIM; it++){
11946: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11947: }
11948: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11949: }
11950:
11951: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11952:
11953: gsl_vector_free(x); /* initial values */
11954: gsl_vector_free(ss); /* inital step size */
11955: for (it=0; it<NDIM; it++){
11956: p[it+1]=gsl_vector_get(sfm->x,it);
11957: fprintf(ficrespow," %.12lf", p[it]);
11958: }
11959: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11960: #endif
11961: #ifdef POWELL
11962: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11963: #endif
1.126 brouard 11964: fclose(ficrespow);
11965:
1.203 brouard 11966: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11967:
11968: for(i=1; i <=NDIM; i++)
11969: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11970: matcov[i][j]=matcov[j][i];
1.126 brouard 11971:
11972: printf("\nCovariance matrix\n ");
1.203 brouard 11973: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11974: for(i=1; i <=NDIM; i++) {
11975: for(j=1;j<=NDIM;j++){
1.220 brouard 11976: printf("%f ",matcov[i][j]);
11977: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11978: }
1.203 brouard 11979: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11980: }
11981:
11982: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11983: for (i=1;i<=NDIM;i++) {
1.126 brouard 11984: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11985: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11986: }
1.126 brouard 11987: lsurv=vector(1,AGESUP);
11988: lpop=vector(1,AGESUP);
11989: tpop=vector(1,AGESUP);
11990: lsurv[agegomp]=100000;
11991:
11992: for (k=agegomp;k<=AGESUP;k++) {
11993: agemortsup=k;
11994: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11995: }
11996:
11997: for (k=agegomp;k<agemortsup;k++)
11998: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11999:
12000: for (k=agegomp;k<agemortsup;k++){
12001: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12002: sumlpop=sumlpop+lpop[k];
12003: }
12004:
12005: tpop[agegomp]=sumlpop;
12006: for (k=agegomp;k<(agemortsup-3);k++){
12007: /* tpop[k+1]=2;*/
12008: tpop[k+1]=tpop[k]-lpop[k];
12009: }
12010:
12011:
12012: printf("\nAge lx qx dx Lx Tx e(x)\n");
12013: for (k=agegomp;k<(agemortsup-2);k++)
12014: 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]);
12015:
12016:
12017: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12018: ageminpar=50;
12019: agemaxpar=100;
1.194 brouard 12020: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12021: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12022: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12023: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12024: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12025: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12026: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12027: }else{
12028: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12029: 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 12030: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12031: }
1.201 brouard 12032: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12033: stepm, weightopt,\
12034: model,imx,p,matcov,agemortsup);
12035:
12036: free_vector(lsurv,1,AGESUP);
12037: free_vector(lpop,1,AGESUP);
12038: free_vector(tpop,1,AGESUP);
1.220 brouard 12039: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12040: free_ivector(dcwave,firstobs,lastobs);
12041: free_vector(agecens,firstobs,lastobs);
12042: free_vector(ageexmed,firstobs,lastobs);
12043: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12044: #ifdef GSL
1.136 brouard 12045: #endif
1.186 brouard 12046: } /* Endof if mle==-3 mortality only */
1.205 brouard 12047: /* Standard */
12048: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12049: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12050: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12051: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12052: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12053: for (k=1; k<=npar;k++)
12054: printf(" %d %8.5f",k,p[k]);
12055: printf("\n");
1.205 brouard 12056: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12057: /* mlikeli uses func not funcone */
1.247 brouard 12058: /* for(i=1;i<nlstate;i++){ */
12059: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12060: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12061: /* } */
1.205 brouard 12062: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12063: }
12064: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12065: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12066: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12067: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12068: }
12069: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12070: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12071: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12072: for (k=1; k<=npar;k++)
12073: printf(" %d %8.5f",k,p[k]);
12074: printf("\n");
12075:
12076: /*--------- results files --------------*/
1.283 brouard 12077: /* 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 12078:
12079:
12080: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12081: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12082: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12083: for(i=1,jk=1; i <=nlstate; i++){
12084: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12085: if (k != i) {
12086: printf("%d%d ",i,k);
12087: fprintf(ficlog,"%d%d ",i,k);
12088: fprintf(ficres,"%1d%1d ",i,k);
12089: for(j=1; j <=ncovmodel; j++){
12090: printf("%12.7f ",p[jk]);
12091: fprintf(ficlog,"%12.7f ",p[jk]);
12092: fprintf(ficres,"%12.7f ",p[jk]);
12093: jk++;
12094: }
12095: printf("\n");
12096: fprintf(ficlog,"\n");
12097: fprintf(ficres,"\n");
12098: }
1.126 brouard 12099: }
12100: }
1.203 brouard 12101: if(mle != 0){
12102: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12103: ftolhess=ftol; /* Usually correct */
1.203 brouard 12104: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12105: 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");
12106: 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");
12107: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12108: for(k=1; k <=(nlstate+ndeath); k++){
12109: if (k != i) {
12110: printf("%d%d ",i,k);
12111: fprintf(ficlog,"%d%d ",i,k);
12112: for(j=1; j <=ncovmodel; j++){
12113: 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]));
12114: 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]));
12115: jk++;
12116: }
12117: printf("\n");
12118: fprintf(ficlog,"\n");
12119: }
12120: }
1.193 brouard 12121: }
1.203 brouard 12122: } /* end of hesscov and Wald tests */
1.225 brouard 12123:
1.203 brouard 12124: /* */
1.126 brouard 12125: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12126: printf("# Scales (for hessian or gradient estimation)\n");
12127: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12128: for(i=1,jk=1; i <=nlstate; i++){
12129: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12130: if (j!=i) {
12131: fprintf(ficres,"%1d%1d",i,j);
12132: printf("%1d%1d",i,j);
12133: fprintf(ficlog,"%1d%1d",i,j);
12134: for(k=1; k<=ncovmodel;k++){
12135: printf(" %.5e",delti[jk]);
12136: fprintf(ficlog," %.5e",delti[jk]);
12137: fprintf(ficres," %.5e",delti[jk]);
12138: jk++;
12139: }
12140: printf("\n");
12141: fprintf(ficlog,"\n");
12142: fprintf(ficres,"\n");
12143: }
1.126 brouard 12144: }
12145: }
12146:
12147: 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 12148: if(mle >= 1) /* To big for the screen */
1.126 brouard 12149: 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");
12150: 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");
12151: /* # 121 Var(a12)\n\ */
12152: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12153: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12154: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12155: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12156: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12157: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12158: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12159:
12160:
12161: /* Just to have a covariance matrix which will be more understandable
12162: even is we still don't want to manage dictionary of variables
12163: */
12164: for(itimes=1;itimes<=2;itimes++){
12165: jj=0;
12166: for(i=1; i <=nlstate; i++){
1.225 brouard 12167: for(j=1; j <=nlstate+ndeath; j++){
12168: if(j==i) continue;
12169: for(k=1; k<=ncovmodel;k++){
12170: jj++;
12171: ca[0]= k+'a'-1;ca[1]='\0';
12172: if(itimes==1){
12173: if(mle>=1)
12174: printf("#%1d%1d%d",i,j,k);
12175: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12176: fprintf(ficres,"#%1d%1d%d",i,j,k);
12177: }else{
12178: if(mle>=1)
12179: printf("%1d%1d%d",i,j,k);
12180: fprintf(ficlog,"%1d%1d%d",i,j,k);
12181: fprintf(ficres,"%1d%1d%d",i,j,k);
12182: }
12183: ll=0;
12184: for(li=1;li <=nlstate; li++){
12185: for(lj=1;lj <=nlstate+ndeath; lj++){
12186: if(lj==li) continue;
12187: for(lk=1;lk<=ncovmodel;lk++){
12188: ll++;
12189: if(ll<=jj){
12190: cb[0]= lk +'a'-1;cb[1]='\0';
12191: if(ll<jj){
12192: if(itimes==1){
12193: if(mle>=1)
12194: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12195: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12196: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12197: }else{
12198: if(mle>=1)
12199: printf(" %.5e",matcov[jj][ll]);
12200: fprintf(ficlog," %.5e",matcov[jj][ll]);
12201: fprintf(ficres," %.5e",matcov[jj][ll]);
12202: }
12203: }else{
12204: if(itimes==1){
12205: if(mle>=1)
12206: printf(" Var(%s%1d%1d)",ca,i,j);
12207: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12208: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12209: }else{
12210: if(mle>=1)
12211: printf(" %.7e",matcov[jj][ll]);
12212: fprintf(ficlog," %.7e",matcov[jj][ll]);
12213: fprintf(ficres," %.7e",matcov[jj][ll]);
12214: }
12215: }
12216: }
12217: } /* end lk */
12218: } /* end lj */
12219: } /* end li */
12220: if(mle>=1)
12221: printf("\n");
12222: fprintf(ficlog,"\n");
12223: fprintf(ficres,"\n");
12224: numlinepar++;
12225: } /* end k*/
12226: } /*end j */
1.126 brouard 12227: } /* end i */
12228: } /* end itimes */
12229:
12230: fflush(ficlog);
12231: fflush(ficres);
1.225 brouard 12232: while(fgets(line, MAXLINE, ficpar)) {
12233: /* If line starts with a # it is a comment */
12234: if (line[0] == '#') {
12235: numlinepar++;
12236: fputs(line,stdout);
12237: fputs(line,ficparo);
12238: fputs(line,ficlog);
1.299 brouard 12239: fputs(line,ficres);
1.225 brouard 12240: continue;
12241: }else
12242: break;
12243: }
12244:
1.209 brouard 12245: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12246: /* ungetc(c,ficpar); */
12247: /* fgets(line, MAXLINE, ficpar); */
12248: /* fputs(line,stdout); */
12249: /* fputs(line,ficparo); */
12250: /* } */
12251: /* ungetc(c,ficpar); */
1.126 brouard 12252:
12253: estepm=0;
1.209 brouard 12254: 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 12255:
12256: if (num_filled != 6) {
12257: 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);
12258: 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);
12259: goto end;
12260: }
12261: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12262: }
12263: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12264: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12265:
1.209 brouard 12266: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12267: if (estepm==0 || estepm < stepm) estepm=stepm;
12268: if (fage <= 2) {
12269: bage = ageminpar;
12270: fage = agemaxpar;
12271: }
12272:
12273: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12274: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12275: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12276:
1.186 brouard 12277: /* Other stuffs, more or less useful */
1.254 brouard 12278: while(fgets(line, MAXLINE, ficpar)) {
12279: /* If line starts with a # it is a comment */
12280: if (line[0] == '#') {
12281: numlinepar++;
12282: fputs(line,stdout);
12283: fputs(line,ficparo);
12284: fputs(line,ficlog);
1.299 brouard 12285: fputs(line,ficres);
1.254 brouard 12286: continue;
12287: }else
12288: break;
12289: }
12290:
12291: 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){
12292:
12293: if (num_filled != 7) {
12294: 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);
12295: 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);
12296: goto end;
12297: }
12298: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12299: 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);
12300: 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);
12301: 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 12302: }
1.254 brouard 12303:
12304: while(fgets(line, MAXLINE, ficpar)) {
12305: /* If line starts with a # it is a comment */
12306: if (line[0] == '#') {
12307: numlinepar++;
12308: fputs(line,stdout);
12309: fputs(line,ficparo);
12310: fputs(line,ficlog);
1.299 brouard 12311: fputs(line,ficres);
1.254 brouard 12312: continue;
12313: }else
12314: break;
1.126 brouard 12315: }
12316:
12317:
12318: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12319: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12320:
1.254 brouard 12321: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12322: if (num_filled != 1) {
12323: 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);
12324: 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);
12325: goto end;
12326: }
12327: printf("pop_based=%d\n",popbased);
12328: fprintf(ficlog,"pop_based=%d\n",popbased);
12329: fprintf(ficparo,"pop_based=%d\n",popbased);
12330: fprintf(ficres,"pop_based=%d\n",popbased);
12331: }
12332:
1.258 brouard 12333: /* Results */
12334: nresult=0;
12335: do{
12336: if(!fgets(line, MAXLINE, ficpar)){
12337: endishere=1;
12338: parameterline=14;
12339: }else if (line[0] == '#') {
12340: /* If line starts with a # it is a comment */
1.254 brouard 12341: numlinepar++;
12342: fputs(line,stdout);
12343: fputs(line,ficparo);
12344: fputs(line,ficlog);
1.299 brouard 12345: fputs(line,ficres);
1.254 brouard 12346: continue;
1.258 brouard 12347: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12348: parameterline=11;
1.296 brouard 12349: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12350: parameterline=12;
12351: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12352: parameterline=13;
12353: else{
12354: parameterline=14;
1.254 brouard 12355: }
1.258 brouard 12356: switch (parameterline){
12357: case 11:
1.296 brouard 12358: 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)){
12359: 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 12360: 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);
12361: 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);
12362: 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);
12363: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12364: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12365: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12366: prvforecast = 1;
12367: }
12368: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.299 brouard 12369: printf("prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12370: fprintf(ficlog,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12371: fprintf(ficres,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12372: prvforecast = 2;
12373: }
12374: else {
12375: 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);
12376: 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);
12377: goto end;
1.258 brouard 12378: }
1.254 brouard 12379: break;
1.258 brouard 12380: case 12:
1.296 brouard 12381: 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)){
12382: 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);
12383: 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);
12384: 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);
12385: 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);
12386: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12387: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12388: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12389: prvbackcast = 1;
12390: }
12391: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.299 brouard 12392: printf("prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.301 ! brouard 12393: fprintf(ficlog,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
! 12394: fprintf(ficres,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12395: prvbackcast = 2;
12396: }
12397: else {
12398: 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);
12399: 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);
12400: goto end;
1.258 brouard 12401: }
1.230 brouard 12402: break;
1.258 brouard 12403: case 13:
12404: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12405: if (num_filled == 0){
12406: resultline[0]='\0';
12407: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12408: 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);
12409: break;
12410: } else if (num_filled != 1){
12411: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12412: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12413: }
12414: nresult++; /* Sum of resultlines */
12415: printf("Result %d: result=%s\n",nresult, resultline);
12416: if(nresult > MAXRESULTLINES){
12417: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12418: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12419: goto end;
12420: }
12421: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12422: fprintf(ficparo,"result: %s\n",resultline);
12423: fprintf(ficres,"result: %s\n",resultline);
12424: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12425: break;
1.258 brouard 12426: case 14:
1.259 brouard 12427: if(ncovmodel >2 && nresult==0 ){
12428: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12429: goto end;
12430: }
1.259 brouard 12431: break;
1.258 brouard 12432: default:
12433: nresult=1;
12434: decoderesult(".",nresult ); /* No covariate */
12435: }
12436: } /* End switch parameterline */
12437: }while(endishere==0); /* End do */
1.126 brouard 12438:
1.230 brouard 12439: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12440: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12441:
12442: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12443: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12444: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12445: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12446: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12447: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12448: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12449: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12450: }else{
1.270 brouard 12451: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12452: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12453: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12454: if(prvforecast==1){
12455: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12456: jprojd=jproj1;
12457: mprojd=mproj1;
12458: anprojd=anproj1;
12459: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12460: jprojf=jproj2;
12461: mprojf=mproj2;
12462: anprojf=anproj2;
12463: } else if(prvforecast == 2){
12464: dateprojd=dateintmean;
12465: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12466: dateprojf=dateintmean+yrfproj;
12467: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12468: }
12469: if(prvbackcast==1){
12470: datebackd=(jback1+12*mback1+365*anback1)/365;
12471: jbackd=jback1;
12472: mbackd=mback1;
12473: anbackd=anback1;
12474: datebackf=(jback2+12*mback2+365*anback2)/365;
12475: jbackf=jback2;
12476: mbackf=mback2;
12477: anbackf=anback2;
12478: } else if(prvbackcast == 2){
12479: datebackd=dateintmean;
12480: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12481: datebackf=dateintmean-yrbproj;
12482: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12483: }
12484:
12485: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12486: }
12487: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12488: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12489: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12490:
1.225 brouard 12491: /*------------ free_vector -------------*/
12492: /* chdir(path); */
1.220 brouard 12493:
1.215 brouard 12494: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12495: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12496: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12497: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12498: free_lvector(num,firstobs,lastobs);
12499: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12500: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12501: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12502: fclose(ficparo);
12503: fclose(ficres);
1.220 brouard 12504:
12505:
1.186 brouard 12506: /* Other results (useful)*/
1.220 brouard 12507:
12508:
1.126 brouard 12509: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12510: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12511: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12512: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12513: fclose(ficrespl);
12514:
12515: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12516: /*#include "hpijx.h"*/
12517: hPijx(p, bage, fage);
1.145 brouard 12518: fclose(ficrespij);
1.227 brouard 12519:
1.220 brouard 12520: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12521: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12522: k=1;
1.126 brouard 12523: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12524:
1.269 brouard 12525: /* Prevalence for each covariate combination in probs[age][status][cov] */
12526: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12527: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12528: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12529: for(k=1;k<=ncovcombmax;k++)
12530: probs[i][j][k]=0.;
1.269 brouard 12531: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12532: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12533: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12534: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12535: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12536: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12537: for(k=1;k<=ncovcombmax;k++)
12538: mobaverages[i][j][k]=0.;
1.219 brouard 12539: mobaverage=mobaverages;
12540: if (mobilav!=0) {
1.235 brouard 12541: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12542: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12543: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12544: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12545: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12546: }
1.269 brouard 12547: } else if (mobilavproj !=0) {
1.235 brouard 12548: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12549: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12550: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12551: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12552: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12553: }
1.269 brouard 12554: }else{
12555: printf("Internal error moving average\n");
12556: fflush(stdout);
12557: exit(1);
1.219 brouard 12558: }
12559: }/* end if moving average */
1.227 brouard 12560:
1.126 brouard 12561: /*---------- Forecasting ------------------*/
1.296 brouard 12562: if(prevfcast==1){
12563: /* /\* if(stepm ==1){*\/ */
12564: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12565: /*This done previously after freqsummary.*/
12566: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12567: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12568:
12569: /* } else if (prvforecast==2){ */
12570: /* /\* if(stepm ==1){*\/ */
12571: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12572: /* } */
12573: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12574: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12575: }
1.269 brouard 12576:
1.296 brouard 12577: /* Prevbcasting */
12578: if(prevbcast==1){
1.219 brouard 12579: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12580: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12581: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12582:
12583: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12584:
12585: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12586:
1.219 brouard 12587: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12588: fclose(ficresplb);
12589:
1.222 brouard 12590: hBijx(p, bage, fage, mobaverage);
12591: fclose(ficrespijb);
1.219 brouard 12592:
1.296 brouard 12593: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12594: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12595: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12596: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12597: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12598: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12599:
12600:
1.269 brouard 12601: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12602:
12603:
1.269 brouard 12604: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12605: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12606: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12607: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12608: } /* end Prevbcasting */
1.268 brouard 12609:
1.186 brouard 12610:
12611: /* ------ Other prevalence ratios------------ */
1.126 brouard 12612:
1.215 brouard 12613: free_ivector(wav,1,imx);
12614: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12615: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12616: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12617:
12618:
1.127 brouard 12619: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12620:
1.201 brouard 12621: strcpy(filerese,"E_");
12622: strcat(filerese,fileresu);
1.126 brouard 12623: if((ficreseij=fopen(filerese,"w"))==NULL) {
12624: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12625: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12626: }
1.208 brouard 12627: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12628: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12629:
12630: pstamp(ficreseij);
1.219 brouard 12631:
1.235 brouard 12632: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12633: if (cptcovn < 1){i1=1;}
12634:
12635: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12636: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12637: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12638: continue;
1.219 brouard 12639: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12640: printf("\n#****** ");
1.225 brouard 12641: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12642: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12643: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12644: }
12645: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12646: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12647: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12648: }
12649: fprintf(ficreseij,"******\n");
1.235 brouard 12650: printf("******\n");
1.219 brouard 12651:
12652: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12653: oldm=oldms;savm=savms;
1.235 brouard 12654: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12655:
1.219 brouard 12656: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12657: }
12658: fclose(ficreseij);
1.208 brouard 12659: printf("done evsij\n");fflush(stdout);
12660: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12661:
1.218 brouard 12662:
1.227 brouard 12663: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12664:
1.201 brouard 12665: strcpy(filerest,"T_");
12666: strcat(filerest,fileresu);
1.127 brouard 12667: if((ficrest=fopen(filerest,"w"))==NULL) {
12668: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12669: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12670: }
1.208 brouard 12671: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12672: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12673: strcpy(fileresstde,"STDE_");
12674: strcat(fileresstde,fileresu);
1.126 brouard 12675: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12676: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12677: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12678: }
1.227 brouard 12679: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12680: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12681:
1.201 brouard 12682: strcpy(filerescve,"CVE_");
12683: strcat(filerescve,fileresu);
1.126 brouard 12684: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12685: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12686: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12687: }
1.227 brouard 12688: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12689: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12690:
1.201 brouard 12691: strcpy(fileresv,"V_");
12692: strcat(fileresv,fileresu);
1.126 brouard 12693: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12694: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12695: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12696: }
1.227 brouard 12697: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12698: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12699:
1.235 brouard 12700: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12701: if (cptcovn < 1){i1=1;}
12702:
12703: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12704: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12705: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12706: continue;
1.242 brouard 12707: printf("\n#****** Result for:");
12708: fprintf(ficrest,"\n#****** Result for:");
12709: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12710: for(j=1;j<=cptcoveff;j++){
12711: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12712: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12713: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12714: }
1.235 brouard 12715: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12716: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12717: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12718: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12719: }
1.208 brouard 12720: fprintf(ficrest,"******\n");
1.227 brouard 12721: fprintf(ficlog,"******\n");
12722: printf("******\n");
1.208 brouard 12723:
12724: fprintf(ficresstdeij,"\n#****** ");
12725: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12726: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12727: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12728: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12729: }
1.235 brouard 12730: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12731: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12732: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12733: }
1.208 brouard 12734: fprintf(ficresstdeij,"******\n");
12735: fprintf(ficrescveij,"******\n");
12736:
12737: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12738: /* pstamp(ficresvij); */
1.225 brouard 12739: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12740: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12741: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12742: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12743: }
1.208 brouard 12744: fprintf(ficresvij,"******\n");
12745:
12746: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12747: oldm=oldms;savm=savms;
1.235 brouard 12748: printf(" cvevsij ");
12749: fprintf(ficlog, " cvevsij ");
12750: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12751: printf(" end cvevsij \n ");
12752: fprintf(ficlog, " end cvevsij \n ");
12753:
12754: /*
12755: */
12756: /* goto endfree; */
12757:
12758: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12759: pstamp(ficrest);
12760:
1.269 brouard 12761: epj=vector(1,nlstate+1);
1.208 brouard 12762: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12763: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12764: cptcod= 0; /* To be deleted */
12765: printf("varevsij vpopbased=%d \n",vpopbased);
12766: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12767: 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 12768: 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 ");
12769: if(vpopbased==1)
12770: 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);
12771: else
1.288 brouard 12772: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12773: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12774: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12775: fprintf(ficrest,"\n");
12776: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12777: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12778: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12779: for(age=bage; age <=fage ;age++){
1.235 brouard 12780: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12781: if (vpopbased==1) {
12782: if(mobilav ==0){
12783: for(i=1; i<=nlstate;i++)
12784: prlim[i][i]=probs[(int)age][i][k];
12785: }else{ /* mobilav */
12786: for(i=1; i<=nlstate;i++)
12787: prlim[i][i]=mobaverage[(int)age][i][k];
12788: }
12789: }
1.219 brouard 12790:
1.227 brouard 12791: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12792: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12793: /* printf(" age %4.0f ",age); */
12794: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12795: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12796: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12797: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12798: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12799: }
12800: epj[nlstate+1] +=epj[j];
12801: }
12802: /* printf(" age %4.0f \n",age); */
1.219 brouard 12803:
1.227 brouard 12804: for(i=1, vepp=0.;i <=nlstate;i++)
12805: for(j=1;j <=nlstate;j++)
12806: vepp += vareij[i][j][(int)age];
12807: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12808: for(j=1;j <=nlstate;j++){
12809: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12810: }
12811: fprintf(ficrest,"\n");
12812: }
1.208 brouard 12813: } /* End vpopbased */
1.269 brouard 12814: free_vector(epj,1,nlstate+1);
1.208 brouard 12815: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12816: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12817: printf("done selection\n");fflush(stdout);
12818: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12819:
1.235 brouard 12820: } /* End k selection */
1.227 brouard 12821:
12822: printf("done State-specific expectancies\n");fflush(stdout);
12823: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12824:
1.288 brouard 12825: /* variance-covariance of forward period prevalence*/
1.269 brouard 12826: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12827:
1.227 brouard 12828:
1.290 brouard 12829: free_vector(weight,firstobs,lastobs);
1.227 brouard 12830: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12831: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12832: free_matrix(anint,1,maxwav,firstobs,lastobs);
12833: free_matrix(mint,1,maxwav,firstobs,lastobs);
12834: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12835: free_ivector(tab,1,NCOVMAX);
12836: fclose(ficresstdeij);
12837: fclose(ficrescveij);
12838: fclose(ficresvij);
12839: fclose(ficrest);
12840: fclose(ficpar);
12841:
12842:
1.126 brouard 12843: /*---------- End : free ----------------*/
1.219 brouard 12844: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12845: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12846: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12847: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12848: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12849: } /* mle==-3 arrives here for freeing */
1.227 brouard 12850: /* endfree:*/
12851: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12852: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12853: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12854: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12855: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12856: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12857: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12858: free_matrix(matcov,1,npar,1,npar);
12859: free_matrix(hess,1,npar,1,npar);
12860: /*free_vector(delti,1,npar);*/
12861: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12862: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12863: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12864: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12865:
12866: free_ivector(ncodemax,1,NCOVMAX);
12867: free_ivector(ncodemaxwundef,1,NCOVMAX);
12868: free_ivector(Dummy,-1,NCOVMAX);
12869: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12870: free_ivector(DummyV,1,NCOVMAX);
12871: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12872: free_ivector(Typevar,-1,NCOVMAX);
12873: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12874: free_ivector(TvarsQ,1,NCOVMAX);
12875: free_ivector(TvarsQind,1,NCOVMAX);
12876: free_ivector(TvarsD,1,NCOVMAX);
12877: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12878: free_ivector(TvarFD,1,NCOVMAX);
12879: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12880: free_ivector(TvarF,1,NCOVMAX);
12881: free_ivector(TvarFind,1,NCOVMAX);
12882: free_ivector(TvarV,1,NCOVMAX);
12883: free_ivector(TvarVind,1,NCOVMAX);
12884: free_ivector(TvarA,1,NCOVMAX);
12885: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12886: free_ivector(TvarFQ,1,NCOVMAX);
12887: free_ivector(TvarFQind,1,NCOVMAX);
12888: free_ivector(TvarVD,1,NCOVMAX);
12889: free_ivector(TvarVDind,1,NCOVMAX);
12890: free_ivector(TvarVQ,1,NCOVMAX);
12891: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12892: free_ivector(Tvarsel,1,NCOVMAX);
12893: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12894: free_ivector(Tposprod,1,NCOVMAX);
12895: free_ivector(Tprod,1,NCOVMAX);
12896: free_ivector(Tvaraff,1,NCOVMAX);
12897: free_ivector(invalidvarcomb,1,ncovcombmax);
12898: free_ivector(Tage,1,NCOVMAX);
12899: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12900: free_ivector(TmodelInvind,1,NCOVMAX);
12901: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12902:
12903: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12904: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12905: fflush(fichtm);
12906: fflush(ficgp);
12907:
1.227 brouard 12908:
1.126 brouard 12909: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12910: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12911: 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 12912: }else{
12913: printf("End of Imach\n");
12914: fprintf(ficlog,"End of Imach\n");
12915: }
12916: printf("See log file on %s\n",filelog);
12917: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12918: /*(void) gettimeofday(&end_time,&tzp);*/
12919: rend_time = time(NULL);
12920: end_time = *localtime(&rend_time);
12921: /* tml = *localtime(&end_time.tm_sec); */
12922: strcpy(strtend,asctime(&end_time));
1.126 brouard 12923: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12924: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12925: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12926:
1.157 brouard 12927: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12928: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12929: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12930: /* printf("Total time was %d uSec.\n", total_usecs);*/
12931: /* if(fileappend(fichtm,optionfilehtm)){ */
12932: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12933: fclose(fichtm);
12934: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12935: fclose(fichtmcov);
12936: fclose(ficgp);
12937: fclose(ficlog);
12938: /*------ End -----------*/
1.227 brouard 12939:
1.281 brouard 12940:
12941: /* Executes gnuplot */
1.227 brouard 12942:
12943: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12944: #ifdef WIN32
1.227 brouard 12945: if (_chdir(pathcd) != 0)
12946: printf("Can't move to directory %s!\n",path);
12947: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12948: #else
1.227 brouard 12949: if(chdir(pathcd) != 0)
12950: printf("Can't move to directory %s!\n", path);
12951: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12952: #endif
1.126 brouard 12953: printf("Current directory %s!\n",pathcd);
12954: /*strcat(plotcmd,CHARSEPARATOR);*/
12955: sprintf(plotcmd,"gnuplot");
1.157 brouard 12956: #ifdef _WIN32
1.126 brouard 12957: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12958: #endif
12959: if(!stat(plotcmd,&info)){
1.158 brouard 12960: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12961: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12962: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12963: }else
12964: strcpy(pplotcmd,plotcmd);
1.157 brouard 12965: #ifdef __unix
1.126 brouard 12966: strcpy(plotcmd,GNUPLOTPROGRAM);
12967: if(!stat(plotcmd,&info)){
1.158 brouard 12968: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12969: }else
12970: strcpy(pplotcmd,plotcmd);
12971: #endif
12972: }else
12973: strcpy(pplotcmd,plotcmd);
12974:
12975: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12976: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 12977: strcpy(pplotcmd,plotcmd);
1.227 brouard 12978:
1.126 brouard 12979: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 12980: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12981: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12982: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 12983: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 12984: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 12985: strcpy(plotcmd,pplotcmd);
12986: }
1.126 brouard 12987: }
1.158 brouard 12988: printf(" Successful, please wait...");
1.126 brouard 12989: while (z[0] != 'q') {
12990: /* chdir(path); */
1.154 brouard 12991: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12992: scanf("%s",z);
12993: /* if (z[0] == 'c') system("./imach"); */
12994: if (z[0] == 'e') {
1.158 brouard 12995: #ifdef __APPLE__
1.152 brouard 12996: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12997: #elif __linux
12998: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12999: #else
1.152 brouard 13000: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13001: #endif
13002: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13003: system(pplotcmd);
1.126 brouard 13004: }
13005: else if (z[0] == 'g') system(plotcmd);
13006: else if (z[0] == 'q') exit(0);
13007: }
1.227 brouard 13008: end:
1.126 brouard 13009: while (z[0] != 'q') {
1.195 brouard 13010: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13011: scanf("%s",z);
13012: }
1.283 brouard 13013: printf("End\n");
1.282 brouard 13014: exit(0);
1.126 brouard 13015: }
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