imach/src/ChangeLog - view

File: [Local Repository] / imach / src / ChangeLog
Revision 1.56: download - view: text, annotated - select for diffs
Sun Aug 21 09:30:43 2022 UTC (22 months ago) by brouard
Branches: MAIN
CVS tags: HEAD

*** empty log message ***

1: 2022-08-21 Nicolas Brouard <brouard@ined.fr> 2: 3: * imach.c (Module): Version 0.99r33 A lot of changes in 4: reassigning covariates: my first idea was that people will always 5: use the first covariate V1 into the model but in fact they are 6: producing data with many covariates and can use an equation model 7: with some of the covariate; it means that in a model V2+V3 instead 8: of codtabm(k,Tvaraff[j]) which calculates for combination k, for 9: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact 10: the equation model is restricted to two variables only (V2, V3) 11: and the combination for V2 should be codtabm(k,1) instead of 12: (codtabm(k,2), and the code should be 13: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been 14: made. All of these should be simplified once a day like we did in 15: hpxij() for example by using precov[nres] which is computed in 16: decoderesult for each nres of each resultline. Loop should be done 17: on the equation model globally by distinguishing only product with 18: age (which are changing with age) and no more on type of 19: covariates, single dummies, single covariates. 20: 21: 2022-08-06 Nicolas Brouard <brouard@ined.fr> 22: 23: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function 24: 25: 2022-08-03 Nicolas Brouard <brouard@ined.fr> 26: 27: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates. 28: 29: 2022-07-25 Brouard Nicolas <brouard@brouard.name> 30: 31: * imach.c (Module): Error cptcovn instead of nsd in bmij (was 32: coredumped, revealed by Feiuno, thank you. 33: 34: 2022-06-02 Brouard Nicolas <brouard@brouard.name> 35: 36: * imach.c (Module): Adding the Wald tests from the log to the main 37: htm for better display of the maximum likelihood estimators. 38: 39: 2022-05-30 Brouard Nicolas <brouard@brouard.name> 40: 41: * imach.c: With products of covariates (age or dummies or 42: quantitatives), estimates of parameters were good but when 43: estimating variances the positions of newly created covariates 44: were wrongly assigned and results were wrong. Thank to Feinua_Sun! 45: 46: 2022-05-24 Nicolas Brouard <brouard@ined.fr> 47: 48: * imach.c (Module): Some attempts to find a bug of wrong estimates 49: of confidencce intervals with product in the equation modelC 50: 51: 2022-05-15 Nicolas Brouard <brouard@ined.fr> 52: 53: * imach.c (Module): Some minor improvements 54: 55: 2022-04-13 Brouard Nicolas <brouard@brouard.name> 56: 57: * imach.c (Module): Adding link to text data files 58: 59: 2022-04-11 Nicolas Brouard <brouard@ined.fr> 60: 61: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed 62: 63: 2022-04-05 Brouard Nicolas <brouard@ined.fr> 64: 65: * imach.c (Module): Fixed covariates (dummy or quantitative) 66: with missing values have never been allowed but are ERRORS and 67: program quits. Standard deviations of fixed covariates were 68: wrongly computed. Mean and standard deviations of time varying 69: covariates are still not computed. 70: 71: 2022-03-17 Brouard Nicolas <brouard@ined.fr> 72: 73: * 99r25 Improvements in fixing discrepancies between covariates 74: missing in result line but in model. 75: 76: *Wed Mar 31 2021 Nicolas Brouard <brouard@ined.fr> 77: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett 78: 79: *2021-03-08 Nicolas Brouard <brouard@ined.fr> 80: 81: * ./ (Module): Fixed bug on result: 82: 83: *2021-02-20 Nicolas Brouard <brouard@ined.fr> 84: 85: * imach.c (Module): Fix bug on quitting after result lines! 86: (Module): Version 0.99r21 87: 88: 2021-02-12 Nicolas Brouard <brouard@ined.fr> 89: 90: * imach.c (Module): The use of a Windows BOM (huge) file is now an error 91: 92: 93: 2021-02-11 Nicolas Brouard <brouard@ined.fr> 94: 95: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed. 96: 97: 2020-02-22 Brouard Nicolas <brouard@ined.fr> 98: 99: * (Module): imach.c Update mle=-3 (for computing Life expectancy 100: and life table from the data without any state) 101: 102: 2019-05-20 Brouard Nicolas <brouard@ined.fr> 103: 104: * version.h: Summary: version 0.99.r19 105: 106: * imach.c: Summary: Projection syntax simplified 107: We can now start projections, forward or backward, from the mean date 108: of inteviews up to or down to a number of years of projection: 109: prevforecast=1 yearsfproj=15.3 mobil_average=0 110: or 111: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0 112: or 113: prevbackcast=1 yearsbproj=12.3 mobil_average=1 114: or 115: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1 116: 117: 2019-05-18 Brouard Nicolas <brouard@ined.fr> 118: 119: * imach.c: Summary: doxygen tex bug 120: 121: 2019-05-16 Brouard Nicolas <brouard@ined.fr> 122: 123: * imach.c: Summary: There was some wrong lines added 124: 125: 2019-05-09 Brouard Nicolas <brouard@ined.fr> 126: 127: * Makefile, imach.c: *** empty log message *** 128: 129: * ChangeLog: Summary: C= 130: 131: * imach.c: Summary: Some updates 132: 133: * imach.c: Summary: Before ncovmax 134: 135: * version.h: *** empty log message *** 136: 137: * imach.c: Summary: 0.99r18 unlimited number of individuals 138: 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. 139: 140: 2018-12-13 Brouard Nicolas <brouard@ined.fr> 141: 142: * imach.c: Summary: Bug for young ages (<-30) will be in r17 143: 144: 2018-12-05 Brouard Nicolas <brouard@ined.fr> 145: 146: * Makefile: Summary: 0.99r16 147: 148: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 149: 150: * imach.c: Summary: Some bugs fixed 151: 152: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 153: 154: * imach.c: Summary: Bug for young ages (<-30) will be in r17 155: 156: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 157: 158: * imach.c: Summary: Some bugs fixed 159: 160: 2018-05-01 Brouard Nicolas <brouard@ined.fr> 161: 162: * imach.c: 163: Summary: Bug fixed by providing frequencies only for non missing covariates 164: 165: 2018-04-27 Brouard Nicolas <brouard@ined.fr> 166: 167: * imach.c: Summary: some minor bugs 168: 169: 2018-04-21 Brouard Nicolas <brouard@ined.fr> 170: 171: * README.txt: *** empty log message *** 172: 173: * imach.c: Summary: Some bugs fixed, valgrind tested 174: 175: 2018-04-20 Brouard Nicolas <brouard@ined.fr> 176: 177: * Makefile, README.txt, setup.iss: Summary: imach 0.99r16 178: 179: * imach.c: 180: Summary: Computing mean and stdeviation of fixed quantitative variables 181: 182: 2018-04-19 Brouard Nicolas <brouard@ined.fr> 183: 184: * imach.c: Summary: Some minor bugs fixed 185: 186: * version.h: *** empty log message *** 187: 188: 2018-02-27 Brouard Nicolas <brouard@ined.fr> 189: 190: * imach.c: *** empty log message *** 191: 192: * imach.c: Summary: Adding second argument for quitting 193: 194: 2018-02-21 Brouard Nicolas <brouard@ined.fr> 195: 196: * Makefile, README.txt, imach.c: Summary: 0.99r15 197: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c 198: 199: 2017-07-20 Brouard Nicolas <brouard@ined.fr> 200: 201: * imach.c: Summary: temporary working 202: 203: 2017-07-19 Brouard Nicolas <brouard@ined.fr> 204: 205: * imach.c: Summary: Bug for mobil_average=0 and prevforecast fixed(?) 206: 207: 2017-07-17 Brouard Nicolas <brouard@ined.fr> 208: 209: * imach.c: Summary: BOM files can be read now 210: 211: 2017-06-30 Brouard Nicolas <brouard@ined.fr> 212: 213: * imach.c: Summary: Graphs improvements 214: 215: * imach.c: Summary: Saito's color 216: 217: 2017-06-29 Brouard Nicolas <brouard@ined.fr> 218: 219: * imach.c: Summary: Version 0.99r14 220: 221: 2017-06-27 Brouard Nicolas <brouard@ined.fr> 222: 223: * imach.c: Summary: More documentation on projections 224: 225: * imach.c: Summary: Color of backprojection changed from 6 to 5(yellow) 226: 227: * imach.c: Summary: Some bug with rint 228: 229: 2017-05-24 Brouard Nicolas <brouard@ined.fr> 230: 231: * imach.c: *** empty log message *** 232: 233: 2017-05-23 Brouard Nicolas <brouard@ined.fr> 234: 235: * imach.c: Summary: Code into subroutine, cleanings 236: 237: 2017-05-18 Brouard Nicolas <brouard@ined.fr> 238: 239: * imach.c: 240: Summary: backprojection and confidence intervals of backprevalence 241: 242: 2017-05-13 Brouard Nicolas <brouard@ined.fr> 243: 244: * imach.c: Summary: temporary save for backprojection 245: 246: * imach.c: Summary: Version 0.99r13 (improvements and bugs fixed) 247: 248: 2017-04-26 Brouard Nicolas <brouard@ined.fr> 249: 250: * imach.c: Summary: imach 0.99r13 Some bugs fixed 251: 252: * imach.c: Summary: Labels in graphs 253: 254: 2017-04-24 Brouard Nicolas <brouard@ined.fr> 255: 256: * imach.c: Summary: to save 257: 258: 2017-04-18 Brouard Nicolas <brouard@ined.fr> 259: 260: * imach.c: *** empty log message *** 261: 262: 2017-04-05 Brouard Nicolas <brouard@ined.fr> 263: 264: * imach.c: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1 265: 266: 2017-04-04 Brouard Nicolas <brouard@ined.fr> 267: 268: * imach.c: Summary: Gnuplot indexations fixed (humm) 269: 270: * imach.c: 271: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3 272: 273: 2017-04-03 Brouard Nicolas <brouard@ined.fr> 274: 275: * imach.c: Summary: Version 0.99r12 276: Some cleanings, conformed with updated documentation. 277: 278: 2017-03-30 Brouard Nicolas <brouard@ined.fr> 279: 280: * version.h: *** empty log message *** 281: 282: * version.h: Summary:0.99r11 283: 284: 2017-03-29 Brouard Nicolas <brouard@ined.fr> 285: 286: * imach.c: Summary: Temp 287: 288: 2017-03-27 Brouard Nicolas <brouard@ined.fr> 289: 290: * imach.c: Summary: Temporary 291: 292: 2017-03-08 Brouard Nicolas <brouard@ined.fr> 293: 294: * imach.c: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed 295: 296: * imach.c: Summary: Fixing data parameter line 297: 298: 2016-12-15 Brouard Nicolas <brouard@ined.fr> 299: 300: * imach.c: Summary: 0.99 in progress 301: 302: 2016-09-15 Brouard Nicolas <brouard@ined.fr> 303: 304: * imach.c: *** empty log message *** 305: 306: * imach.c: Summary: not working 307: 308: 2016-09-08 Brouard Nicolas <brouard@ined.fr> 309: 310: * imach.c: Summary: continue 311: 312: 2016-09-07 Brouard Nicolas <brouard@ined.fr> 313: 314: * imach.c: Summary: Starting values from frequencies 315: 316: * imach.c: *** empty log message *** 317: 318: 2016-09-02 Brouard Nicolas <brouard@ined.fr> 319: 320: * imach.c: *** empty log message *** 321: 322: 2016-08-30 Brouard Nicolas <brouard@ined.fr> 323: 324: * imach.c: Summary: Fixing a lots 325: 326: 2016-08-29 Brouard Nicolas <brouard@ined.fr> 327: 328: * imach.c: Summary: gnuplot problem in Back projection to fix 329: 330: * imach.c: Summary: Better 331: 332: 2016-08-26 Brouard Nicolas <brouard@ined.fr> 333: 334: * imach.c: 335: Summary: Improvement in Powell output in order to copy and paste 336: 337: * imach.c: Summary: Starting tests of 0.99 338: 339: * imach.c: Summary: to valgrind 340: 341: 2016-08-25 Brouard Nicolas <brouard@ined.fr> 342: 343: * imach.c: *** empty log message *** 344: 345: 2016-08-23 Brouard Nicolas <brouard@ined.fr> 346: 347: * imach.c: *** empty log message *** 348: 349: * imach.c: Summary: not working 350: 351: 2016-08-22 Brouard Nicolas <brouard@ined.fr> 352: 353: * imach.c: Summary: not working 354: 355: * imach.c: Summary: Not working 356: 357: 2016-07-23 Brouard Nicolas <brouard@ined.fr> 358: 359: * imach.c: Summary: Completing for func too 360: 361: 2016-07-22 Brouard Nicolas <brouard@ined.fr> 362: 363: * imach.c: Summary: Fixing some arrays, still debugging 364: 365: 2016-07-21 Brouard Nicolas <brouard@ined.fr> 366: 367: * imach.c: 368: Summary: 0.99 working (more or less) for Asian Workshop on multitate methods 369: 370: 2016-07-12 Brouard Nicolas <brouard@ined.fr> 371: 372: * imach.c: Summary: temp 373: 374: * imach.c: Summary: saving but not running 375: 376: 2016-07-01 Brouard Nicolas <brouard@ined.fr> 377: 378: * imach.c: Summary: Fixes 379: 380: 2016-02-19 Brouard Nicolas <brouard@ined.fr> 381: 382: * imach-0.98r.c: Summary: Kind of 0.98r? series, starting with r7 383: 384: * setup.iss, Makefile, version.h: *** empty log message *** 385: 386: * imach.c: Summary: temporary 387: 388: 2016-02-17 Brouard Nicolas <brouard@ined.fr> 389: 390: * imach.c: Summary: Probably last 0.98 stable version 0.98r6 391: 392: 2016-02-16 Brouard Nicolas <brouard@ined.fr> 393: 394: * imach.c: Summary: minor bug 395: 396: * imach.c: Summary: 0.99r2 397: 398: 2016-02-15 Brouard Nicolas <brouard@ined.fr> 399: 400: * imach.c: *** empty log message *** 401: 402: 2016-02-12 Brouard Nicolas <brouard@ined.fr> 403: 404: * imach.c: Summary: 0.99 Back projections 405: 406: 2015-12-23 Brouard Nicolas <brouard@ined.fr> 407: 408: * imach.c: Summary: Experimental backcast 409: 410: 2015-12-18 Brouard Nicolas <brouard@ined.fr> 411: 412: * imach.c: Summary: 0.98r4 Warning and status=-2 413: 414: Version 0.98r4 is now: 415: - displaying an error when status is -1, date of interview unknown and date of death known; 416: - permitting a status -2 when the vital status is unknown at a known date of right truncation. 417: Older changes concerning s=-2, dating from 2005 have been supersed. 418: 419: * Makefile: *** empty log message *** 420: 421: 2015-12-16 Brouard Nicolas <brouard@ined.fr> 422: 423: * Makefile: *** empty log message *** 424: 425: * imach.c: Summary: 0.98r4 working 426: 427: * Makefile: Summary: linuxrpm added 428: 429: * Makefile: *** empty log message *** 430: 431: * imach.c: Summary: temporary not working 432: 433: 2015-12-11 Brouard Nicolas <brouard@ined.fr> 434: 435: * imach.c: Summary: 0.98r4 436: 437: 2015-12-08 Brouard Nicolas <brouard@ined.fr> 438: 439: * Makefile: *** empty log message *** 440: 441: 2015-11-21 Brouard Nicolas <brouard@ined.fr> 442: 443: * Makefile: 444: Summary: VBoxManage change from execute --image to run --exe and --wait-exit doesn't exist any more 445: 446: * imach.c: Summary: minor typo 447: 448: * imach.c: Summary: 0.98r3 with some graph of projected cross-sectional 449: 450: Author: Nicolas Brouard 451: 452: 2015-11-18 Brouard Nicolas <brouard@ined.fr> 453: 454: * imach.c: Summary: Start working on projected prevalences 455: 456: 2015-11-17 Brouard Nicolas <brouard@ined.fr> 457: 458: * imach.c: Summary: Adding ftolpl parameter 459: Author: N Brouard 460: 461: We had difficulties to get smoothed confidence intervals. It was due 462: to the period prevalence which wasn't computed accurately. The inner 463: parameter ftolpl is now an outer parameter of the .imach parameter 464: file after estepm. If ftolpl is small 1.e-4 and estepm too, 465: computation are long. 466: 467: * version.h: *** empty log message *** 468: 469: * imach.c: Summary: temporary 470: 471: 2015-10-27 Brouard Nicolas <brouard@ined.fr> 472: 473: * Makefile: *** empty log message *** 474: 475: * setup.iss.in: Summary: some new subdirs 476: 477: * imach.c: *** empty log message *** 478: 479: * setup.iss.in: 480: Summary: Adding gnuplot 5.1 with bin and etc for pango fonts 481: 482: 2015-10-24 Brouard Nicolas <brouard@ined.fr> 483: 484: * imach.c: *** empty log message *** 485: 486: 2015-10-23 Brouard Nicolas <brouard@ined.fr> 487: 488: * imach.c: 489: Summary: 0.98r3 some clarification for graphs on likelihood contributions 490: 491: 2015-10-01 Brouard Nicolas <brouard@ined.fr> 492: 493: * imach.c: Summary: Some new graphs of contribution to likelihood 494: 495: 2015-09-30 Brouard Nicolas <brouard@ined.fr> 496: 497: * Makefile: 498: Summary: Version read from version.h, itself read with cmake 499: 500: * imach.c: Summary: looking at better estimation of the hessian 501: 502: Also a better criteria for convergence to the period prevalence And 503: therefore adding the number of years needed to converge. (The 504: prevalence in any alive state shold sum to one 505: 506: * Makefile: *** empty log message *** 507: 508: 2015-09-22 Brouard Nicolas <brouard@ined.fr> 509: 510: * imach.c: 511: Summary: Adding some overall graph on contribution to likelihood. Might change 512: 513: 2015-09-15 Brouard Nicolas <brouard@ined.fr> 514: 515: * Makefile: *** empty log message *** 516: 517: * imach.c: Summary: 0.98r0 518: 519: - Some new graphs like survival functions 520: - Some bugs fixed like model=1+age+V2. 521: 522: * ChangeLog, Makefile: Summary: 0.98r0 523: 524: * CMakeLists.txt: Summary: Some tests for setup.iss 525: 526: 2015-09-15 Nicolas J Brouard <nbrouard@tugault.ined.fr> 527: 528: * (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc 529: 530: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local> 531: 532: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line. 533: 534: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr> 535: 536: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih) 537: 538: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local> 539: 540: * imach.c (Module): Merging 1.61 to 1.162 541: 542: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr> 543: 544: * prevlim.h (Module): 545: 546: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local> 547: 548: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested... 549: (Module): Version 0.98nR Running ok, but output format still only works for three covariates. 550: 551: 2010-04-29 brouard <brouard@ined.fr> 552: 553: * imach.c (Module): Checking covariates for more complex models 554: than V1+V2. A lot of change to be done. Unstable. 555: 556: 2010-04-26 brouard <brouard@ined.fr> 557: 558: * imach.c (Module): merging some libgsl code. Fixing computation 559: of likelione (using inter/intrapolation if mle = 0) in order to 560: get same likelihood as if mle=1. 561: Some cleaning of code and comments added. 562: 563: 2009-10-29 brouard <brouard@ined.fr> 564: 565: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code. 566: 567: 2006-06-30 Brouard Nicolas <brouard@localhost> 568: 569: * imach.c (Module): Clarifications on computing e.j 570: 571: 2006-04-28 Brouard Nicolas <brouard@localhost> 572: 573: * imach.c (Module): Yes the sum of survivors was wrong since 574: imach-114 because nhstepm was no more computed in the age 575: loop. Now we define nhstepma in the age loop. 576: (Module): In order to speed up (in case of numerous covariates) we 577: compute health expectancies (without variances) in a first step 578: and then all the health expectancies with variances or standard 579: deviation (needs data from the Hessian matrices) which slows the 580: computation. 581: In the future we should be able to stop the program is only health 582: expectancies and graph are needed without standard deviations. 583: 584: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr> 585: 586: * imach.iss (Module): 587: 588: * imach.c (Module): bugs in evsi + population file added in htm output file 589: 590: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr> 591: 592: * imach.c (Module): 593: 594: 2006-03-20 Brouard Nicolas <brouard@localhost> 595: 596: * imach.c (Module): <title> changed, corresponds to .htm file 597: name. <head> headers where missing. 598: 599: * Makefile (Module): 0.98g 600: 601: * imach.c (Module): Weights can have a decimal point as for 602: English (a comma might work with a correct LC_NUMERIC environment, 603: otherwise the weight is truncated). 604: Modification of warning when the covariates values are not 0 or 605: 1. 606: Version 0.98g 607: 608: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr> 609: 610: * imach.c (Module): Comments concerning covariates added 611: 612: * noreg-setup.iss (Module): Version 0.98f 613: 614: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup 615: 616: * imach.c (Module): refinements in the computation of lli if 617: status=-2 in order to have more reliable computation if stepm is 618: not 1 month. Version 0.98f 619: 620: 2006-03-15 Brouard Nicolas <brouard@localhost> 621: 622: * imach.c (Module): Bug if status = -2, the loglikelihood was 623: computed as likelihood omitting the logarithm. Version O.98e 624: 625: 2006-03-14 Brouard Nicolas <brouard@localhost> 626: 627: * imach.c (Module): varevsij Comments added explaining the second 628: table of variances if popbased=1 . 629: (Module): Covariances of eij, ekl added, graphs fixed, new html link. 630: (Module): Function pstamp added 631: (Module): Version 0.98d 632: 633: 2006-03-06 Brouard Nicolas <brouard@localhost> 634: 635: * imach.c (Module): Variance-covariance wrong links and 636: varian-covariance of ej. is needed (Saito). 637: 638: 2006-02-27 Brouard Nicolas <brouard@localhost> 639: 640: * imach.c (Module): One freematrix added in mlikeli! 641: 642: 2006-02-26 Brouard Nicolas <brouard@localhost> 643: 644: * imach.c (Module): Some improvements in processing parameter 645: filename with strsep. 646: 647: 2006-02-24 Brouard Nicolas <brouard@localhost> 648: 649: * imach.c (Module): Memory leaks checks with valgrind and: 650: datafile was not closed, some imatrix were not freed and on matrix 651: allocation too. 652: 653: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr> 654: 655: * strsep.c (Module): Inclusion because not in mingw32!!! 656: 657: * setup.iss.in (Module): IMACHSETUPVERSION added 658: 659: * noreg-setup.iss (Module): added to cvs tree 660: 661: * Makefile (Module): IMACHSETUPVERSION added 662: 663: 2006-01-30 BROUARD Nicolas <brouard@localhost> 664: 665: * Makefile (Module, Module): Latest Makefile 666: 667: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe 668: 669: 2006-01-27 BROUARD Nicolas <brouard@localhost> 670: 671: * noreg-setup.iss (Module): ADDED 672: 673: 2006-01-24 BROUARD Nicolas <brouard@localhost> 674: 675: * imach.c (Module): Comments (lines starting with a #) are allowed in data. 676: 677: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr> 678: 679: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg 680: 681: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X 682: 683: 2005-09-30 <lievre@PC_00595> 684: 685: * imach.c (Module): sump fixed, loop imx fixed, and simplifications. 686: (Module): If the status is missing at the last wave but we know 687: that the person is alive, then we can code his/her status as -2 688: (instead of missing=-1 in earlier versions) and his/her 689: contributions to the likelihood is 1 - Prob of dying from last 690: health status (= 1-p13= p11+p12 in the easiest case of somebody in 691: the healthy state at last known wave). Version is 0.98 692: 693: 2004-05-20 Brouard Nicolas <brouard@localhost> 694: 695: * imach.c (Repository): 696: Agnes added a direct estimation of mortality (without the need of 697: computing period prevalence and differential mortality). Thus here 698: is version 0.97a which has been distributed to some people at 699: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of 700: mortality using covariates is not done today. Estimating direct 701: mortality is a very different process because it doesn't need 702: interpolation because it is easy to get the lx from the force of 703: the mortality mux in the simplest case as for a Gompertz (log mux 704: = a + b*x . But we have been able to incorporate the new code 705: within former imach program (0.96d) without deteriorating too much 706: the understanding of the program. 707: 708: 2003-06-25 <brouard@BROUARD> 709: 710: * imach.c (Module): On windows (cygwin) function asctime_r doesn't 711: exist so I changed back to asctime which exists. 712: (Module): Version 0.96b 713: 714: 2003-06-24 <brouard@BROUARD> 715: 716: * imach.c (Module): Some bugs corrected for windows. Also, when 717: mle=-1 a template is output in file "or"mypar.txt with the design 718: of the covariance matrix to be input. 719: 720: 2003-06-25 Brouard Nicolas <brouard@homebrou> 721: 722: * imach.c (Repository): Duplicated warning errors corrected. 723: (Repository): Elapsed time after each iteration is now output. It 724: helps to forecast when convergence will be reached. Elapsed time 725: is stamped in powell. We created a new html file for the graphs 726: concerning matrix of covariance. It has extension -cov.htm. 727: 728: 2003-06-23 Brouard Nicolas <brouard@homebrou> 729: 730: * 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. 731: 732: 2003-06-18 Brouard Nicolas <brouard@homebrou> 733: 734: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96 735: 736: 2003-06-17 Brouard Nicolas <brouard@homebrou> 737: 738: 739: * imach.c (Repository): Check when date of death was earlier that 740: current date of interview. It may happen when the death was just 741: prior to the death. In this case, dh was negative and likelihood 742: was wrong (infinity). We still send an "Error" but patch by 743: assuming that the date of death was just one stepm after the 744: interview. 745: (Repository): Because some people have very long ID (first column) 746: we changed int to long in num[] and we added a new lvector for 747: memory allocation. But we also truncated to 8 characters (left 748: truncation) 749: 750: (Repository): No more line truncation errors. 751: 752: 2003-06-14 Brouard Nicolas <brouard@homebrou> 753: 754: * imach.c (Repository): Add of a routine likelione (likelihood 755: only once) which prints on a text file 756: (ilk) the contributions to the likelihood for each 757: individual/wave. 758: 759: 2003-06-13 Brouard Nicolas <brouard@homebrou> 760: * imach.c (Repository): Replace "freqsummary" at a correct 761: place. It differs from routine "prevalence" which may be called 762: many times. Probs is memory consuming and must be used with 763: parcimony. 764: 765: 2003-06-17 <brouard@BROUARD> 766: 767: * timeval.h (Module): Added included file to make use of 768: gettimeofday working on win32 with cygwin. 769: 770: 2003-05-16 Brouard Nicolas <brouard@brouard> 771: * imach.c (Module): Insert of a warning if the delay between two 772: waves is negative. 773: 774: * imach.c (Module): 775: 776: 2003-05-03 <brouard@BROUARD> 777: 778: * imach.c: Some cleaning 779: 780: 2003-03-28 <brouard@BROUARD> 781: * imach.c (Module): In version up to 0.92 likelihood was computed 782: as if date of death was unknown. Death was treated as any other 783: health state: the date of the interview describes the actual state 784: and not the date of a change in health state. The former idea was 785: to consider that at each interview the state was recorded 786: (healthy, disable or death) and IMaCh was corrected; but when we 787: introduced the exact date of death then we should have modified 788: the contribution of an exact death to the likelihood. This new 789: contribution is smaller and very dependent of the step unit 790: stepm. It is no more the probability to die between last interview 791: and month of death but the probability to survive from last 792: interview up to one month before death multiplied by the 793: probability to die within a month. Thanks to Chris 794: Jackson for correcting this bug. Former versions increased 795: mortality artificially. The bad side is that we add another loop 796: which slows down the processing. The difference can be up to 10% 797: lower mortality. 798: 799: 2002-05-30 <brouard@BROUARD> 800: * 801: * imach.c (Module): Add correlation matrix of one-step 802: probabilities and covariance matrix 803: 804: 805: 806: 807: 808: ======= 809: >>>>>>> 1.55 810: 2022-07-23 Nicolas Brouard <brouard@ined.fr> 811: 812: * r29 W and not sqrt(Wald) 813: 814: 2022-07-22 Nicolas Brouard <brouard@ined.fr> 815: 816: * imach.c (Module): Output of Wald test in the htm file and not only in the log. 817: 818: 2022-05-30 Brouard Nicolas <brouard@brouard.name> 819: 820: * imach.c: With products of covariates (age or dummies or 821: quantitatives), estimates of parameters were good but when 822: estimating variances the positions of newly created covariates 823: were wrongly assigned and results were wrong. Thank to Feinua_Sun! 824: 825: 2022-05-24 Nicolas Brouard <brouard@ined.fr> 826: 827: * imach.c (Module): Some attempts to find a bug of wrong estimates 828: of confidencce intervals with product in the equation modelC 829: 830: 2022-05-15 Nicolas Brouard <brouard@ined.fr> 831: 832: * imach.c (Module): Some minor improvements 833: 834: 2022-04-13 Brouard Nicolas <brouard@brouard.name> 835: 836: * imach.c (Module): Adding link to text data files 837: 838: 2022-04-11 Nicolas Brouard <brouard@ined.fr> 839: 840: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed 841: 842: 2022-04-05 Brouard Nicolas <brouard@ined.fr> 843: 844: * imach.c (Module): Fixed covariates (dummy or quantitative) 845: with missing values have never been allowed but are ERRORS and 846: program quits. Standard deviations of fixed covariates were 847: wrongly computed. Mean and standard deviations of time varying 848: covariates are still not computed. 849: 850: 2022-03-17 Brouard Nicolas <brouard@ined.fr> 851: 852: * 99r25 Improvements in fixing discrepancies between covariates 853: missing in result line but in model. 854: 855: *Wed Mar 31 2021 Nicolas Brouard <brouard@ined.fr> 856: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett 857: 858: *2021-03-08 Nicolas Brouard <brouard@ined.fr> 859: 860: * ./ (Module): Fixed bug on result: 861: 862: *2021-02-20 Nicolas Brouard <brouard@ined.fr> 863: 864: * imach.c (Module): Fix bug on quitting after result lines! 865: (Module): Version 0.99r21 866: 867: 2021-02-12 Nicolas Brouard <brouard@ined.fr> 868: 869: * imach.c (Module): The use of a Windows BOM (huge) file is now an error 870: 871: 872: 2021-02-11 Nicolas Brouard <brouard@ined.fr> 873: 874: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed. 875: 876: 2020-02-22 Brouard Nicolas <brouard@ined.fr> 877: 878: * (Module): imach.c Update mle=-3 (for computing Life expectancy 879: and life table from the data without any state) 880: 881: 2019-05-20 Brouard Nicolas <brouard@ined.fr> 882: 883: * version.h: Summary: version 0.99.r19 884: 885: * imach.c: Summary: Projection syntax simplified 886: We can now start projections, forward or backward, from the mean date 887: of inteviews up to or down to a number of years of projection: 888: prevforecast=1 yearsfproj=15.3 mobil_average=0 889: or 890: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0 891: or 892: prevbackcast=1 yearsbproj=12.3 mobil_average=1 893: or 894: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1 895: 896: 2019-05-18 Brouard Nicolas <brouard@ined.fr> 897: 898: * imach.c: Summary: doxygen tex bug 899: 900: 2019-05-16 Brouard Nicolas <brouard@ined.fr> 901: 902: * imach.c: Summary: There was some wrong lines added 903: 904: 2019-05-09 Brouard Nicolas <brouard@ined.fr> 905: 906: * Makefile, imach.c: *** empty log message *** 907: 908: * ChangeLog: Summary: C= 909: 910: * imach.c: Summary: Some updates 911: 912: * imach.c: Summary: Before ncovmax 913: 914: * version.h: *** empty log message *** 915: 916: * imach.c: Summary: 0.99r18 unlimited number of individuals 917: 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. 918: 919: 2018-12-13 Brouard Nicolas <brouard@ined.fr> 920: 921: * imach.c: Summary: Bug for young ages (<-30) will be in r17 922: 923: 2018-12-05 Brouard Nicolas <brouard@ined.fr> 924: 925: * Makefile: Summary: 0.99r16 926: 927: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 928: 929: * imach.c: Summary: Some bugs fixed 930: 931: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 932: 933: * imach.c: Summary: Bug for young ages (<-30) will be in r17 934: 935: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 936: 937: * imach.c: Summary: Some bugs fixed 938: 939: 2018-05-01 Brouard Nicolas <brouard@ined.fr> 940: 941: * imach.c: 942: Summary: Bug fixed by providing frequencies only for non missing covariates 943: 944: 2018-04-27 Brouard Nicolas <brouard@ined.fr> 945: 946: * imach.c: Summary: some minor bugs 947: 948: 2018-04-21 Brouard Nicolas <brouard@ined.fr> 949: 950: * README.txt: *** empty log message *** 951: 952: * imach.c: Summary: Some bugs fixed, valgrind tested 953: 954: 2018-04-20 Brouard Nicolas <brouard@ined.fr> 955: 956: * Makefile, README.txt, setup.iss: Summary: imach 0.99r16 957: 958: * imach.c: 959: Summary: Computing mean and stdeviation of fixed quantitative variables 960: 961: 2018-04-19 Brouard Nicolas <brouard@ined.fr> 962: 963: * imach.c: Summary: Some minor bugs fixed 964: 965: * version.h: *** empty log message *** 966: 967: 2018-02-27 Brouard Nicolas <brouard@ined.fr> 968: 969: * imach.c: *** empty log message *** 970: 971: * imach.c: Summary: Adding second argument for quitting 972: 973: 2018-02-21 Brouard Nicolas <brouard@ined.fr> 974: 975: * Makefile, README.txt, imach.c: Summary: 0.99r15 976: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c 977: 978: 2017-07-20 Brouard Nicolas <brouard@ined.fr> 979: 980: * imach.c: Summary: temporary working 981: 982: 2017-07-19 Brouard Nicolas <brouard@ined.fr> 983: 984: * imach.c: Summary: Bug for mobil_average=0 and prevforecast fixed(?) 985: 986: 2017-07-17 Brouard Nicolas <brouard@ined.fr> 987: 988: * imach.c: Summary: BOM files can be read now 989: 990: 2017-06-30 Brouard Nicolas <brouard@ined.fr> 991: 992: * imach.c: Summary: Graphs improvements 993: 994: * imach.c: Summary: Saito's color 995: 996: 2017-06-29 Brouard Nicolas <brouard@ined.fr> 997: 998: * imach.c: Summary: Version 0.99r14 999: 1000: 2017-06-27 Brouard Nicolas <brouard@ined.fr> 1001: 1002: * imach.c: Summary: More documentation on projections 1003: 1004: * imach.c: Summary: Color of backprojection changed from 6 to 5(yellow) 1005: 1006: * imach.c: Summary: Some bug with rint 1007: 1008: 2017-05-24 Brouard Nicolas <brouard@ined.fr> 1009: 1010: * imach.c: *** empty log message *** 1011: 1012: 2017-05-23 Brouard Nicolas <brouard@ined.fr> 1013: 1014: * imach.c: Summary: Code into subroutine, cleanings 1015: 1016: 2017-05-18 Brouard Nicolas <brouard@ined.fr> 1017: 1018: * imach.c: 1019: Summary: backprojection and confidence intervals of backprevalence 1020: 1021: 2017-05-13 Brouard Nicolas <brouard@ined.fr> 1022: 1023: * imach.c: Summary: temporary save for backprojection 1024: 1025: * imach.c: Summary: Version 0.99r13 (improvements and bugs fixed) 1026: 1027: 2017-04-26 Brouard Nicolas <brouard@ined.fr> 1028: 1029: * imach.c: Summary: imach 0.99r13 Some bugs fixed 1030: 1031: * imach.c: Summary: Labels in graphs 1032: 1033: 2017-04-24 Brouard Nicolas <brouard@ined.fr> 1034: 1035: * imach.c: Summary: to save 1036: 1037: 2017-04-18 Brouard Nicolas <brouard@ined.fr> 1038: 1039: * imach.c: *** empty log message *** 1040: 1041: 2017-04-05 Brouard Nicolas <brouard@ined.fr> 1042: 1043: * imach.c: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1 1044: 1045: 2017-04-04 Brouard Nicolas <brouard@ined.fr> 1046: 1047: * imach.c: Summary: Gnuplot indexations fixed (humm) 1048: 1049: * imach.c: 1050: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3 1051: 1052: 2017-04-03 Brouard Nicolas <brouard@ined.fr> 1053: 1054: * imach.c: Summary: Version 0.99r12 1055: Some cleanings, conformed with updated documentation. 1056: 1057: 2017-03-30 Brouard Nicolas <brouard@ined.fr> 1058: 1059: * version.h: *** empty log message *** 1060: 1061: * version.h: Summary:0.99r11 1062: 1063: 2017-03-29 Brouard Nicolas <brouard@ined.fr> 1064: 1065: * imach.c: Summary: Temp 1066: 1067: 2017-03-27 Brouard Nicolas <brouard@ined.fr> 1068: 1069: * imach.c: Summary: Temporary 1070: 1071: 2017-03-08 Brouard Nicolas <brouard@ined.fr> 1072: 1073: * imach.c: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed 1074: 1075: * imach.c: Summary: Fixing data parameter line 1076: 1077: 2016-12-15 Brouard Nicolas <brouard@ined.fr> 1078: 1079: * imach.c: Summary: 0.99 in progress 1080: 1081: 2016-09-15 Brouard Nicolas <brouard@ined.fr> 1082: 1083: * imach.c: *** empty log message *** 1084: 1085: * imach.c: Summary: not working 1086: 1087: 2016-09-08 Brouard Nicolas <brouard@ined.fr> 1088: 1089: * imach.c: Summary: continue 1090: 1091: 2016-09-07 Brouard Nicolas <brouard@ined.fr> 1092: 1093: * imach.c: Summary: Starting values from frequencies 1094: 1095: * imach.c: *** empty log message *** 1096: 1097: 2016-09-02 Brouard Nicolas <brouard@ined.fr> 1098: 1099: * imach.c: *** empty log message *** 1100: 1101: 2016-08-30 Brouard Nicolas <brouard@ined.fr> 1102: 1103: * imach.c: Summary: Fixing a lots 1104: 1105: 2016-08-29 Brouard Nicolas <brouard@ined.fr> 1106: 1107: * imach.c: Summary: gnuplot problem in Back projection to fix 1108: 1109: * imach.c: Summary: Better 1110: 1111: 2016-08-26 Brouard Nicolas <brouard@ined.fr> 1112: 1113: * imach.c: 1114: Summary: Improvement in Powell output in order to copy and paste 1115: 1116: * imach.c: Summary: Starting tests of 0.99 1117: 1118: * imach.c: Summary: to valgrind 1119: 1120: 2016-08-25 Brouard Nicolas <brouard@ined.fr> 1121: 1122: * imach.c: *** empty log message *** 1123: 1124: 2016-08-23 Brouard Nicolas <brouard@ined.fr> 1125: 1126: * imach.c: *** empty log message *** 1127: 1128: * imach.c: Summary: not working 1129: 1130: 2016-08-22 Brouard Nicolas <brouard@ined.fr> 1131: 1132: * imach.c: Summary: not working 1133: 1134: * imach.c: Summary: Not working 1135: 1136: 2016-07-23 Brouard Nicolas <brouard@ined.fr> 1137: 1138: * imach.c: Summary: Completing for func too 1139: 1140: 2016-07-22 Brouard Nicolas <brouard@ined.fr> 1141: 1142: * imach.c: Summary: Fixing some arrays, still debugging 1143: 1144: 2016-07-21 Brouard Nicolas <brouard@ined.fr> 1145: 1146: * imach.c: 1147: Summary: 0.99 working (more or less) for Asian Workshop on multitate methods 1148: 1149: 2016-07-12 Brouard Nicolas <brouard@ined.fr> 1150: 1151: * imach.c: Summary: temp 1152: 1153: * imach.c: Summary: saving but not running 1154: 1155: 2016-07-01 Brouard Nicolas <brouard@ined.fr> 1156: 1157: * imach.c: Summary: Fixes 1158: 1159: 2016-02-19 Brouard Nicolas <brouard@ined.fr> 1160: 1161: * imach-0.98r.c: Summary: Kind of 0.98r? series, starting with r7 1162: 1163: * setup.iss, Makefile, version.h: *** empty log message *** 1164: 1165: * imach.c: Summary: temporary 1166: 1167: 2016-02-17 Brouard Nicolas <brouard@ined.fr> 1168: 1169: * imach.c: Summary: Probably last 0.98 stable version 0.98r6 1170: 1171: 2016-02-16 Brouard Nicolas <brouard@ined.fr> 1172: 1173: * imach.c: Summary: minor bug 1174: 1175: * imach.c: Summary: 0.99r2 1176: 1177: 2016-02-15 Brouard Nicolas <brouard@ined.fr> 1178: 1179: * imach.c: *** empty log message *** 1180: 1181: 2016-02-12 Brouard Nicolas <brouard@ined.fr> 1182: 1183: * imach.c: Summary: 0.99 Back projections 1184: 1185: 2015-12-23 Brouard Nicolas <brouard@ined.fr> 1186: 1187: * imach.c: Summary: Experimental backcast 1188: 1189: 2015-12-18 Brouard Nicolas <brouard@ined.fr> 1190: 1191: * imach.c: Summary: 0.98r4 Warning and status=-2 1192: 1193: Version 0.98r4 is now: 1194: - displaying an error when status is -1, date of interview unknown and date of death known; 1195: - permitting a status -2 when the vital status is unknown at a known date of right truncation. 1196: Older changes concerning s=-2, dating from 2005 have been supersed. 1197: 1198: * Makefile: *** empty log message *** 1199: 1200: 2015-12-16 Brouard Nicolas <brouard@ined.fr> 1201: 1202: * Makefile: *** empty log message *** 1203: 1204: * imach.c: Summary: 0.98r4 working 1205: 1206: * Makefile: Summary: linuxrpm added 1207: 1208: * Makefile: *** empty log message *** 1209: 1210: * imach.c: Summary: temporary not working 1211: 1212: 2015-12-11 Brouard Nicolas <brouard@ined.fr> 1213: 1214: * imach.c: Summary: 0.98r4 1215: 1216: 2015-12-08 Brouard Nicolas <brouard@ined.fr> 1217: 1218: * Makefile: *** empty log message *** 1219: 1220: 2015-11-21 Brouard Nicolas <brouard@ined.fr> 1221: 1222: * Makefile: 1223: Summary: VBoxManage change from execute --image to run --exe and --wait-exit doesn't exist any more 1224: 1225: * imach.c: Summary: minor typo 1226: 1227: * imach.c: Summary: 0.98r3 with some graph of projected cross-sectional 1228: 1229: Author: Nicolas Brouard 1230: 1231: 2015-11-18 Brouard Nicolas <brouard@ined.fr> 1232: 1233: * imach.c: Summary: Start working on projected prevalences 1234: 1235: 2015-11-17 Brouard Nicolas <brouard@ined.fr> 1236: 1237: * imach.c: Summary: Adding ftolpl parameter 1238: Author: N Brouard 1239: 1240: We had difficulties to get smoothed confidence intervals. It was due 1241: to the period prevalence which wasn't computed accurately. The inner 1242: parameter ftolpl is now an outer parameter of the .imach parameter 1243: file after estepm. If ftolpl is small 1.e-4 and estepm too, 1244: computation are long. 1245: 1246: * version.h: *** empty log message *** 1247: 1248: * imach.c: Summary: temporary 1249: 1250: 2015-10-27 Brouard Nicolas <brouard@ined.fr> 1251: 1252: * Makefile: *** empty log message *** 1253: 1254: * setup.iss.in: Summary: some new subdirs 1255: 1256: * imach.c: *** empty log message *** 1257: 1258: * setup.iss.in: 1259: Summary: Adding gnuplot 5.1 with bin and etc for pango fonts 1260: 1261: 2015-10-24 Brouard Nicolas <brouard@ined.fr> 1262: 1263: * imach.c: *** empty log message *** 1264: 1265: 2015-10-23 Brouard Nicolas <brouard@ined.fr> 1266: 1267: * imach.c: 1268: Summary: 0.98r3 some clarification for graphs on likelihood contributions 1269: 1270: 2015-10-01 Brouard Nicolas <brouard@ined.fr> 1271: 1272: * imach.c: Summary: Some new graphs of contribution to likelihood 1273: 1274: 2015-09-30 Brouard Nicolas <brouard@ined.fr> 1275: 1276: * Makefile: 1277: Summary: Version read from version.h, itself read with cmake 1278: 1279: * imach.c: Summary: looking at better estimation of the hessian 1280: 1281: Also a better criteria for convergence to the period prevalence And 1282: therefore adding the number of years needed to converge. (The 1283: prevalence in any alive state shold sum to one 1284: 1285: * Makefile: *** empty log message *** 1286: 1287: 2015-09-22 Brouard Nicolas <brouard@ined.fr> 1288: 1289: * imach.c: 1290: Summary: Adding some overall graph on contribution to likelihood. Might change 1291: 1292: 2015-09-15 Brouard Nicolas <brouard@ined.fr> 1293: 1294: * Makefile: *** empty log message *** 1295: 1296: * imach.c: Summary: 0.98r0 1297: 1298: - Some new graphs like survival functions 1299: - Some bugs fixed like model=1+age+V2. 1300: 1301: * ChangeLog, Makefile: Summary: 0.98r0 1302: 1303: * CMakeLists.txt: Summary: Some tests for setup.iss 1304: 1305: 2015-09-15 Nicolas J Brouard <nbrouard@tugault.ined.fr> 1306: 1307: * (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc 1308: 1309: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local> 1310: 1311: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line. 1312: 1313: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr> 1314: 1315: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih) 1316: 1317: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local> 1318: 1319: * imach.c (Module): Merging 1.61 to 1.162 1320: 1321: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr> 1322: 1323: * prevlim.h (Module): 1324: 1325: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local> 1326: 1327: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested... 1328: (Module): Version 0.98nR Running ok, but output format still only works for three covariates. 1329: 1330: 2010-04-29 brouard <brouard@ined.fr> 1331: 1332: * imach.c (Module): Checking covariates for more complex models 1333: than V1+V2. A lot of change to be done. Unstable. 1334: 1335: 2010-04-26 brouard <brouard@ined.fr> 1336: 1337: * imach.c (Module): merging some libgsl code. Fixing computation 1338: of likelione (using inter/intrapolation if mle = 0) in order to 1339: get same likelihood as if mle=1. 1340: Some cleaning of code and comments added. 1341: 1342: 2009-10-29 brouard <brouard@ined.fr> 1343: 1344: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code. 1345: 1346: 2006-06-30 Brouard Nicolas <brouard@localhost> 1347: 1348: * imach.c (Module): Clarifications on computing e.j 1349: 1350: 2006-04-28 Brouard Nicolas <brouard@localhost> 1351: 1352: * imach.c (Module): Yes the sum of survivors was wrong since 1353: imach-114 because nhstepm was no more computed in the age 1354: loop. Now we define nhstepma in the age loop. 1355: (Module): In order to speed up (in case of numerous covariates) we 1356: compute health expectancies (without variances) in a first step 1357: and then all the health expectancies with variances or standard 1358: deviation (needs data from the Hessian matrices) which slows the 1359: computation. 1360: In the future we should be able to stop the program is only health 1361: expectancies and graph are needed without standard deviations. 1362: 1363: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr> 1364: 1365: * imach.iss (Module): 1366: 1367: * imach.c (Module): bugs in evsi + population file added in htm output file 1368: 1369: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr> 1370: 1371: * imach.c (Module): 1372: 1373: 2006-03-20 Brouard Nicolas <brouard@localhost> 1374: 1375: * imach.c (Module): <title> changed, corresponds to .htm file 1376: name. <head> headers where missing. 1377: 1378: * Makefile (Module): 0.98g 1379: 1380: * imach.c (Module): Weights can have a decimal point as for 1381: English (a comma might work with a correct LC_NUMERIC environment, 1382: otherwise the weight is truncated). 1383: Modification of warning when the covariates values are not 0 or 1384: 1. 1385: Version 0.98g 1386: 1387: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr> 1388: 1389: * imach.c (Module): Comments concerning covariates added 1390: 1391: * noreg-setup.iss (Module): Version 0.98f 1392: 1393: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup 1394: 1395: * imach.c (Module): refinements in the computation of lli if 1396: status=-2 in order to have more reliable computation if stepm is 1397: not 1 month. Version 0.98f 1398: 1399: 2006-03-15 Brouard Nicolas <brouard@localhost> 1400: 1401: * imach.c (Module): Bug if status = -2, the loglikelihood was 1402: computed as likelihood omitting the logarithm. Version O.98e 1403: 1404: 2006-03-14 Brouard Nicolas <brouard@localhost> 1405: 1406: * imach.c (Module): varevsij Comments added explaining the second 1407: table of variances if popbased=1 . 1408: (Module): Covariances of eij, ekl added, graphs fixed, new html link. 1409: (Module): Function pstamp added 1410: (Module): Version 0.98d 1411: 1412: 2006-03-06 Brouard Nicolas <brouard@localhost> 1413: 1414: * imach.c (Module): Variance-covariance wrong links and 1415: varian-covariance of ej. is needed (Saito). 1416: 1417: 2006-02-27 Brouard Nicolas <brouard@localhost> 1418: 1419: * imach.c (Module): One freematrix added in mlikeli! 1420: 1421: 2006-02-26 Brouard Nicolas <brouard@localhost> 1422: 1423: * imach.c (Module): Some improvements in processing parameter 1424: filename with strsep. 1425: 1426: 2006-02-24 Brouard Nicolas <brouard@localhost> 1427: 1428: * imach.c (Module): Memory leaks checks with valgrind and: 1429: datafile was not closed, some imatrix were not freed and on matrix 1430: allocation too. 1431: 1432: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr> 1433: 1434: * strsep.c (Module): Inclusion because not in mingw32!!! 1435: 1436: * setup.iss.in (Module): IMACHSETUPVERSION added 1437: 1438: * noreg-setup.iss (Module): added to cvs tree 1439: 1440: * Makefile (Module): IMACHSETUPVERSION added 1441: 1442: 2006-01-30 BROUARD Nicolas <brouard@localhost> 1443: 1444: * Makefile (Module, Module): Latest Makefile 1445: 1446: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe 1447: 1448: 2006-01-27 BROUARD Nicolas <brouard@localhost> 1449: 1450: * noreg-setup.iss (Module): ADDED 1451: 1452: 2006-01-24 BROUARD Nicolas <brouard@localhost> 1453: 1454: * imach.c (Module): Comments (lines starting with a #) are allowed in data. 1455: 1456: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr> 1457: 1458: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg 1459: 1460: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X 1461: 1462: 2005-09-30 <lievre@PC_00595> 1463: 1464: * imach.c (Module): sump fixed, loop imx fixed, and simplifications. 1465: (Module): If the status is missing at the last wave but we know 1466: that the person is alive, then we can code his/her status as -2 1467: (instead of missing=-1 in earlier versions) and his/her 1468: contributions to the likelihood is 1 - Prob of dying from last 1469: health status (= 1-p13= p11+p12 in the easiest case of somebody in 1470: the healthy state at last known wave). Version is 0.98 1471: 1472: 2004-05-20 Brouard Nicolas <brouard@localhost> 1473: 1474: * imach.c (Repository): 1475: Agnes added a direct estimation of mortality (without the need of 1476: computing period prevalence and differential mortality). Thus here 1477: is version 0.97a which has been distributed to some people at 1478: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of 1479: mortality using covariates is not done today. Estimating direct 1480: mortality is a very different process because it doesn't need 1481: interpolation because it is easy to get the lx from the force of 1482: the mortality mux in the simplest case as for a Gompertz (log mux 1483: = a + b*x . But we have been able to incorporate the new code 1484: within former imach program (0.96d) without deteriorating too much 1485: the understanding of the program. 1486: 1487: 2003-06-25 <brouard@BROUARD> 1488: 1489: * imach.c (Module): On windows (cygwin) function asctime_r doesn't 1490: exist so I changed back to asctime which exists. 1491: (Module): Version 0.96b 1492: 1493: 2003-06-24 <brouard@BROUARD> 1494: 1495: * imach.c (Module): Some bugs corrected for windows. Also, when 1496: mle=-1 a template is output in file "or"mypar.txt with the design 1497: of the covariance matrix to be input. 1498: 1499: 2003-06-25 Brouard Nicolas <brouard@homebrou> 1500: 1501: * imach.c (Repository): Duplicated warning errors corrected. 1502: (Repository): Elapsed time after each iteration is now output. It 1503: helps to forecast when convergence will be reached. Elapsed time 1504: is stamped in powell. We created a new html file for the graphs 1505: concerning matrix of covariance. It has extension -cov.htm. 1506: 1507: 2003-06-23 Brouard Nicolas <brouard@homebrou> 1508: 1509: * 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. 1510: 1511: 2003-06-18 Brouard Nicolas <brouard@homebrou> 1512: 1513: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96 1514: 1515: 2003-06-17 Brouard Nicolas <brouard@homebrou> 1516: 1517: 1518: * imach.c (Repository): Check when date of death was earlier that 1519: current date of interview. It may happen when the death was just 1520: prior to the death. In this case, dh was negative and likelihood 1521: was wrong (infinity). We still send an "Error" but patch by 1522: assuming that the date of death was just one stepm after the 1523: interview. 1524: (Repository): Because some people have very long ID (first column) 1525: we changed int to long in num[] and we added a new lvector for 1526: memory allocation. But we also truncated to 8 characters (left 1527: truncation) 1528: 1529: (Repository): No more line truncation errors. 1530: 1531: 2003-06-14 Brouard Nicolas <brouard@homebrou> 1532: 1533: * imach.c (Repository): Add of a routine likelione (likelihood 1534: only once) which prints on a text file 1535: (ilk) the contributions to the likelihood for each 1536: individual/wave. 1537: 1538: 2003-06-13 Brouard Nicolas <brouard@homebrou> 1539: * imach.c (Repository): Replace "freqsummary" at a correct 1540: place. It differs from routine "prevalence" which may be called 1541: many times. Probs is memory consuming and must be used with 1542: parcimony. 1543: 1544: 2003-06-17 <brouard@BROUARD> 1545: 1546: * timeval.h (Module): Added included file to make use of 1547: gettimeofday working on win32 with cygwin. 1548: 1549: 2003-05-16 Brouard Nicolas <brouard@brouard> 1550: * imach.c (Module): Insert of a warning if the delay between two 1551: waves is negative. 1552: 1553: * imach.c (Module): 1554: 1555: 2003-05-03 <brouard@BROUARD> 1556: 1557: * imach.c: Some cleaning 1558: 1559: 2003-03-28 <brouard@BROUARD> 1560: * imach.c (Module): In version up to 0.92 likelihood was computed 1561: as if date of death was unknown. Death was treated as any other 1562: health state: the date of the interview describes the actual state 1563: and not the date of a change in health state. The former idea was 1564: to consider that at each interview the state was recorded 1565: (healthy, disable or death) and IMaCh was corrected; but when we 1566: introduced the exact date of death then we should have modified 1567: the contribution of an exact death to the likelihood. This new 1568: contribution is smaller and very dependent of the step unit 1569: stepm. It is no more the probability to die between last interview 1570: and month of death but the probability to survive from last 1571: interview up to one month before death multiplied by the 1572: probability to die within a month. Thanks to Chris 1573: Jackson for correcting this bug. Former versions increased 1574: mortality artificially. The bad side is that we add another loop 1575: which slows down the processing. The difference can be up to 10% 1576: lower mortality. 1577: 1578: 2002-05-30 <brouard@BROUARD> 1579: * 1580: * imach.c (Module): Add correlation matrix of one-step 1581: probabilities and covariance matrix 1582: 1583: 1584: 1585: 1586: