imach/src/ChangeLog - view

File: [Local Repository] / imach / src / ChangeLog
Revision 1.38: download - view: text, annotated - select for diffs
Thu May 9 15:06:14 2019 UTC (5 years, 5 months ago) by brouard
Branches: MAIN
CVS tags: HEAD

Summary: C=

1: 2018-05-02 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 2: 3: * imach.c: Summary: Some bugs fixed 4: 5: 2018-05-01 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 6: 7: * imach.c: 8: Summary: Bug fixed by providing frequencies only for non missing covariates 9: 10: 2018-04-27 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 11: 12: * imach.c: Summary: some minor bugs 13: 14: 2018-04-21 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 15: 16: * README.txt: *** empty log message *** 17: 18: * imach.c: Summary: Some bugs fixed, valgrind tested 19: 20: 2018-04-20 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 21: 22: * Makefile, README.txt, setup.iss: Summary: imach 0.99r16 23: 24: * imach.c: 25: Summary: Computing mean and stdeviation of fixed quantitative variables 26: 27: 2018-04-19 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 28: 29: * imach.c: Summary: Some minor bugs fixed 30: 31: * version.h: *** empty log message *** 32: 33: 2018-02-27 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 34: 35: * imach.c: *** empty log message *** 36: 37: * imach.c: Summary: Adding second argument for quitting 38: 39: 2018-02-21 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 40: 41: * Makefile, README.txt, imach.c: Summary: 0.99r15 42: 43: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c 44: 45: 2017-07-20 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 46: 47: * imach.c: Summary: temporary working 48: 49: 2017-07-19 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 50: 51: * imach.c: Summary: Bug for mobil_average=0 and prevforecast fixed(?) 52: 53: 2017-07-17 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 54: 55: * imach.c: Summary: BOM files can be read now 56: 57: 2017-06-30 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 58: 59: * imach.c: Summary: Graphs improvements 60: 61: * imach.c: Summary: Saito's color 62: 63: 2017-06-29 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 64: 65: * imach.c: Summary: Version 0.99r14 66: 67: 2017-06-27 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 68: 69: * imach.c: Summary: More documentation on projections 70: 71: * imach.c: Summary: Color of backprojection changed from 6 to 5(yellow) 72: 73: * imach.c: Summary: Some bug with rint 74: 75: 2017-05-24 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 76: 77: * imach.c: *** empty log message *** 78: 79: 2017-05-23 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 80: 81: * imach.c: Summary: Code into subroutine, cleanings 82: 83: 2017-05-18 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 84: 85: * imach.c: 86: Summary: backprojection and confidence intervals of backprevalence 87: 88: 2017-05-13 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 89: 90: * imach.c: Summary: temporary save for backprojection 91: 92: * imach.c: Summary: Version 0.99r13 (improvements and bugs fixed) 93: 94: 2017-04-26 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 95: 96: * imach.c: Summary: imach 0.99r13 Some bugs fixed 97: 98: * imach.c: Summary: Labels in graphs 99: 100: 2017-04-24 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 101: 102: * imach.c: Summary: to save 103: 104: 2017-04-18 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 105: 106: * imach.c: *** empty log message *** 107: 108: 2017-04-05 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 109: 110: * imach.c: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1 111: 112: 2017-04-04 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 113: 114: * imach.c: Summary: Gnuplot indexations fixed (humm) 115: 116: * imach.c: 117: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3 118: 119: 2017-04-03 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 120: 121: * imach.c: Summary: Version 0.99r12 122: 123: Some cleanings, conformed with updated documentation. 124: 125: 2017-03-30 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 126: 127: * version.h: *** empty log message *** 128: 129: * version.h: Summary:0.99r11 130: 131: 2017-03-29 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 132: 133: * imach.c: Summary: Temp 134: 135: 2017-03-27 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 136: 137: * imach.c: Summary: Temporary 138: 139: 2017-03-08 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 140: 141: * imach.c: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed 142: 143: * imach.c: Summary: Fixing data parameter line 144: 145: 2016-12-15 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 146: 147: * imach.c: Summary: 0.99 in progress 148: 149: 2016-09-15 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 150: 151: * imach.c: *** empty log message *** 152: 153: * imach.c: Summary: not working 154: 155: 2016-09-08 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 156: 157: * imach.c: Summary: continue 158: 159: 2016-09-07 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 160: 161: * imach.c: Summary: Starting values from frequencies 162: 163: * imach.c: *** empty log message *** 164: 165: 2016-09-02 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 166: 167: * imach.c: *** empty log message *** 168: 169: 2016-08-30 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 170: 171: * imach.c: Summary: Fixing a lots 172: 173: 2016-08-29 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 174: 175: * imach.c: Summary: gnuplot problem in Back projection to fix 176: 177: * imach.c: Summary: Better 178: 179: 2016-08-26 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 180: 181: * imach.c: 182: Summary: Improvement in Powell output in order to copy and paste 183: 184: Author: 185: 186: * imach.c: Summary: Starting tests of 0.99 187: 188: * imach.c: Summary: to valgrind 189: 190: 2016-08-25 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 191: 192: * imach.c: *** empty log message *** 193: 194: 2016-08-23 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 195: 196: * imach.c: *** empty log message *** 197: 198: * imach.c: Summary: not working 199: 200: 2016-08-22 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 201: 202: * imach.c: Summary: not working 203: 204: * imach.c: Summary: Not working 205: 206: 2016-07-23 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 207: 208: * imach.c: Summary: Completing for func too 209: 210: 2016-07-22 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 211: 212: * imach.c: Summary: Fixing some arrays, still debugging 213: 214: 2016-07-21 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 215: 216: * imach.c: 217: Summary: 0.99 working (more or less) for Asian Workshop on multitate methods 218: 219: 2016-07-12 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 220: 221: * imach.c: Summary: temp 222: 223: * imach.c: Summary: saving but not running 224: 225: 2016-07-01 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 226: 227: * imach.c: Summary: Fixes 228: 229: 2016-02-19 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 230: 231: * imach-0.98r.c: Summary: Kind of 0.98r? series, starting with r7 232: 233: * setup.iss, Makefile, version.h: *** empty log message *** 234: 235: * imach.c: Summary: temporary 236: 237: 2016-02-17 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 238: 239: * imach.c: Summary: Probably last 0.98 stable version 0.98r6 240: 241: 2016-02-16 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 242: 243: * imach.c: Summary: minor bug 244: 245: * imach.c: Summary: 0.99r2 246: 247: 2016-02-15 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 248: 249: * imach.c: *** empty log message *** 250: 251: 2016-02-12 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 252: 253: * imach.c: Summary: 0.99 Back projections 254: 255: 2015-12-23 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 256: 257: * imach.c: Summary: Experimental backcast 258: 259: 2015-12-18 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 260: 261: * imach.c: Summary: 0.98r4 Warning and status=-2 262: 263: Version 0.98r4 is now: 264: - displaying an error when status is -1, date of interview unknown and date of death known; 265: - permitting a status -2 when the vital status is unknown at a known date of right truncation. 266: Older changes concerning s=-2, dating from 2005 have been supersed. 267: 268: * Makefile: *** empty log message *** 269: 270: 2015-12-16 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 271: 272: * Makefile: *** empty log message *** 273: 274: * imach.c: Summary: 0.98r4 working 275: 276: * Makefile: Summary: linuxrpm added 277: 278: * Makefile: *** empty log message *** 279: 280: * imach.c: Summary: temporary not working 281: 282: 2015-12-11 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 283: 284: * imach.c: Summary: 0.98r4 285: 286: 2015-12-08 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 287: 288: * Makefile: *** empty log message *** 289: 290: 2015-11-21 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 291: 292: * Makefile: 293: Summary: VBoxManage change from execute --image to run --exe and --wait-exit doesn't exist any more 294: 295: * imach.c: Summary: minor typo 296: 297: * imach.c: Summary: 0.98r3 with some graph of projected cross-sectional 298: 299: Author: Nicolas Brouard 300: 301: 2015-11-18 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 302: 303: * imach.c: Summary: Start working on projected prevalences 304: 305: 2015-11-17 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 306: 307: * imach.c: Summary: Adding ftolpl parameter 308: Author: N Brouard 309: 310: We had difficulties to get smoothed confidence intervals. It was due 311: to the period prevalence which wasn't computed accurately. The inner 312: parameter ftolpl is now an outer parameter of the .imach parameter 313: file after estepm. If ftolpl is small 1.e-4 and estepm too, 314: computation are long. 315: 316: * version.h: *** empty log message *** 317: 318: * imach.c: Summary: temporary 319: 320: 2015-10-27 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 321: 322: * Makefile: *** empty log message *** 323: 324: * setup.iss.in: Summary: some new subdirs 325: 326: * imach.c: *** empty log message *** 327: 328: * setup.iss.in: 329: Summary: Adding gnuplot 5.1 with bin and etc for pango fonts 330: 331: 2015-10-24 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 332: 333: * imach.c: *** empty log message *** 334: 335: 2015-10-23 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 336: 337: * imach.c: 338: Summary: 0.98r3 some clarification for graphs on likelihood contributions 339: 340: 2015-10-01 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 341: 342: * imach.c: Summary: Some new graphs of contribution to likelihood 343: 344: 2015-09-30 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 345: 346: * Makefile: 347: Summary: Version read from version.h, itself read with cmake 348: 349: * imach.c: Summary: looking at better estimation of the hessian 350: 351: Also a better criteria for convergence to the period prevalence And 352: therefore adding the number of years needed to converge. (The 353: prevalence in any alive state shold sum to one 354: 355: * Makefile: *** empty log message *** 356: 357: 2015-09-22 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 358: 359: * imach.c: 360: Summary: Adding some overall graph on contribution to likelihood. Might change 361: 362: 2015-09-15 Brouard Nicolas <brouard@brouard-antoine.homelinux.com> 363: 364: * Makefile: *** empty log message *** 365: 366: * imach.c: Summary: 0.98r0 367: 368: - Some new graphs like suvival functions 369: - Some bugs fixed like model=1+age+V2. 370: 371: * ChangeLog, Makefile: Summary: 0.98r0 372: 373: * CMakeLists.txt: Summary: Some tests for setup.iss 374: 375: 2015-09-15 Nicolas J Brouard <nbrouard@tugault.ined.fr> 376: 377: * (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc 378: 379: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local> 380: 381: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line. 382: 383: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr> 384: 385: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih) 386: 387: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local> 388: 389: * imach.c (Module): Merging 1.61 to 1.162 390: 391: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr> 392: 393: * prevlim.h (Module): 394: 395: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local> 396: 397: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested... 398: (Module): Version 0.98nR Running ok, but output format still only works for three covariates. 399: 400: 2010-04-29 brouard <brouard@ined.fr> 401: 402: * imach.c (Module): Checking covariates for more complex models 403: than V1+V2. A lot of change to be done. Unstable. 404: 405: 2010-04-26 brouard <brouard@ined.fr> 406: 407: * imach.c (Module): merging some libgsl code. Fixing computation 408: of likelione (using inter/intrapolation if mle = 0) in order to 409: get same likelihood as if mle=1. 410: Some cleaning of code and comments added. 411: 412: 2009-10-29 brouard <brouard@ined.fr> 413: 414: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code. 415: 416: 2006-06-30 Brouard Nicolas <brouard@localhost> 417: 418: * imach.c (Module): Clarifications on computing e.j 419: 420: 2006-04-28 Brouard Nicolas <brouard@localhost> 421: 422: * imach.c (Module): Yes the sum of survivors was wrong since 423: imach-114 because nhstepm was no more computed in the age 424: loop. Now we define nhstepma in the age loop. 425: (Module): In order to speed up (in case of numerous covariates) we 426: compute health expectancies (without variances) in a first step 427: and then all the health expectancies with variances or standard 428: deviation (needs data from the Hessian matrices) which slows the 429: computation. 430: In the future we should be able to stop the program is only health 431: expectancies and graph are needed without standard deviations. 432: 433: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr> 434: 435: * imach.iss (Module): 436: 437: * imach.c (Module): bugs in evsi + population file added in htm output file 438: 439: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr> 440: 441: * imach.c (Module): 442: 443: 2006-03-20 Brouard Nicolas <brouard@localhost> 444: 445: * imach.c (Module): <title> changed, corresponds to .htm file 446: name. <head> headers where missing. 447: 448: * Makefile (Module): 0.98g 449: 450: * imach.c (Module): Weights can have a decimal point as for 451: English (a comma might work with a correct LC_NUMERIC environment, 452: otherwise the weight is truncated). 453: Modification of warning when the covariates values are not 0 or 454: 1. 455: Version 0.98g 456: 457: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr> 458: 459: * imach.c (Module): Comments concerning covariates added 460: 461: * noreg-setup.iss (Module): Version 0.98f 462: 463: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup 464: 465: * imach.c (Module): refinements in the computation of lli if 466: status=-2 in order to have more reliable computation if stepm is 467: not 1 month. Version 0.98f 468: 469: 2006-03-15 Brouard Nicolas <brouard@localhost> 470: 471: * imach.c (Module): Bug if status = -2, the loglikelihood was 472: computed as likelihood omitting the logarithm. Version O.98e 473: 474: 2006-03-14 Brouard Nicolas <brouard@localhost> 475: 476: * imach.c (Module): varevsij Comments added explaining the second 477: table of variances if popbased=1 . 478: (Module): Covariances of eij, ekl added, graphs fixed, new html link. 479: (Module): Function pstamp added 480: (Module): Version 0.98d 481: 482: 2006-03-06 Brouard Nicolas <brouard@localhost> 483: 484: * imach.c (Module): Variance-covariance wrong links and 485: varian-covariance of ej. is needed (Saito). 486: 487: 2006-02-27 Brouard Nicolas <brouard@localhost> 488: 489: * imach.c (Module): One freematrix added in mlikeli! 490: 491: 2006-02-26 Brouard Nicolas <brouard@localhost> 492: 493: * imach.c (Module): Some improvements in processing parameter 494: filename with strsep. 495: 496: 2006-02-24 Brouard Nicolas <brouard@localhost> 497: 498: * imach.c (Module): Memory leaks checks with valgrind and: 499: datafile was not closed, some imatrix were not freed and on matrix 500: allocation too. 501: 502: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr> 503: 504: * strsep.c (Module): Inclusion because not in mingw32!!! 505: 506: * setup.iss.in (Module): IMACHSETUPVERSION added 507: 508: * noreg-setup.iss (Module): added to cvs tree 509: 510: * Makefile (Module): IMACHSETUPVERSION added 511: 512: 2006-01-30 BROUARD Nicolas <brouard@localhost> 513: 514: * Makefile (Module, Module): Latest Makefile 515: 516: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe 517: 518: 2006-01-27 BROUARD Nicolas <brouard@localhost> 519: 520: * noreg-setup.iss (Module): ADDED 521: 522: 2006-01-24 BROUARD Nicolas <brouard@localhost> 523: 524: * imach.c (Module): Comments (lines starting with a #) are allowed in data. 525: 526: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr> 527: 528: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg 529: 530: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X 531: 532: 2005-09-30 <lievre@PC_00595> 533: 534: * imach.c (Module): sump fixed, loop imx fixed, and simplifications. 535: (Module): If the status is missing at the last wave but we know 536: that the person is alive, then we can code his/her status as -2 537: (instead of missing=-1 in earlier versions) and his/her 538: contributions to the likelihood is 1 - Prob of dying from last 539: health status (= 1-p13= p11+p12 in the easiest case of somebody in 540: the healthy state at last known wave). Version is 0.98 541: 542: 2004-05-20 Brouard Nicolas <brouard@localhost> 543: 544: * imach.c (Repository): 545: Agnes added a direct estimation of mortality (without the need of 546: computing period prevalence and differential mortality). Thus here 547: is version 0.97a which has been distributed to some people at 548: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of 549: mortality using covariates is not done today. Estimating direct 550: mortality is a very different process because it doesn't need 551: interpolation because it is easy to get the lx from the force of 552: the mortality mux in the simplest case as for a Gompertz (log mux 553: = a + b*x . But we have been able to incorporate the new code 554: within former imach program (0.96d) without deteriorating too much 555: the understanding of the program. 556: 557: 2003-06-25 <brouard@BROUARD> 558: 559: * imach.c (Module): On windows (cygwin) function asctime_r doesn't 560: exist so I changed back to asctime which exists. 561: (Module): Version 0.96b 562: 563: 2003-06-24 <brouard@BROUARD> 564: 565: * imach.c (Module): Some bugs corrected for windows. Also, when 566: mle=-1 a template is output in file "or"mypar.txt with the design 567: of the covariance matrix to be input. 568: 569: 2003-06-25 Brouard Nicolas <brouard@homebrou> 570: 571: * imach.c (Repository): Duplicated warning errors corrected. 572: (Repository): Elapsed time after each iteration is now output. It 573: helps to forecast when convergence will be reached. Elapsed time 574: is stamped in powell. We created a new html file for the graphs 575: concerning matrix of covariance. It has extension -cov.htm. 576: 577: 2003-06-23 Brouard Nicolas <brouard@homebrou> 578: 579: * 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. 580: 581: 2003-06-18 Brouard Nicolas <brouard@homebrou> 582: 583: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96 584: 585: 2003-06-17 Brouard Nicolas <brouard@homebrou> 586: 587: 588: * imach.c (Repository): Check when date of death was earlier that 589: current date of interview. It may happen when the death was just 590: prior to the death. In this case, dh was negative and likelihood 591: was wrong (infinity). We still send an "Error" but patch by 592: assuming that the date of death was just one stepm after the 593: interview. 594: (Repository): Because some people have very long ID (first column) 595: we changed int to long in num[] and we added a new lvector for 596: memory allocation. But we also truncated to 8 characters (left 597: truncation) 598: 599: (Repository): No more line truncation errors. 600: 601: 2003-06-14 Brouard Nicolas <brouard@homebrou> 602: 603: * imach.c (Repository): Add of a routine likelione (likelihood 604: only once) which prints on a text file 605: (ilk) the contributions to the likelihood for each 606: individual/wave. 607: 608: 2003-06-13 Brouard Nicolas <brouard@homebrou> 609: * imach.c (Repository): Replace "freqsummary" at a correct 610: place. It differs from routine "prevalence" which may be called 611: many times. Probs is memory consuming and must be used with 612: parcimony. 613: 614: 2003-06-17 <brouard@BROUARD> 615: 616: * timeval.h (Module): Added included file to make use of 617: gettimeofday working on win32 with cygwin. 618: 619: 2003-05-16 Brouard Nicolas <brouard@brouard> 620: * imach.c (Module): Insert of a warning if the delay between two 621: waves is negative. 622: 623: * imach.c (Module): 624: 625: 2003-05-03 <brouard@BROUARD> 626: 627: * imach.c: Some cleaning 628: 629: 2003-03-28 <brouard@BROUARD> 630: * imach.c (Module): In version up to 0.92 likelihood was computed 631: as if date of death was unknown. Death was treated as any other 632: health state: the date of the interview describes the actual state 633: and not the date of a change in health state. The former idea was 634: to consider that at each interview the state was recorded 635: (healthy, disable or death) and IMaCh was corrected; but when we 636: introduced the exact date of death then we should have modified 637: the contribution of an exact death to the likelihood. This new 638: contribution is smaller and very dependent of the step unit 639: stepm. It is no more the probability to die between last interview 640: and month of death but the probability to survive from last 641: interview up to one month before death multiplied by the 642: probability to die within a month. Thanks to Chris 643: Jackson for correcting this bug. Former versions increased 644: mortality artificially. The bad side is that we add another loop 645: 646: 647: which slows down the processing. The difference can be up to 10% 648: 649: 650: lower mortality. 651: 652: 653: 654: 655: 656: 2002-05-30 <brouard@BROUARD> 657: 658: 659: 660: 661: 662: * 663: 664: 665: 666: 667: 668: * imach.c (Module): Add correlation matrix of one-step 669: 670: 671: probabilities and covariance matrix 672: 673: 674: 675: 676: