imach/src/ChangeLog - view

File: [Local Repository] / imach / src / ChangeLog
Revision 1.74: download - view: text, annotated - select for diffs
Mon Jul 8 14:25:11 2024 UTC (2 weeks ago) by brouard
Branches: MAIN
CVS tags: HEAD

Summary: 0.99s7

1: 2024-07-08 Nicolas Brouard <brouard@ined.fr> 2: 3: * imach.c (Module): Some bug fixes: in drawings when age*age is 4: included in the model as well as with quantitative variables. 5: 6: 2024-07-02 Nicolas Brouard <brouard@ined.fr> 7: 8: * imach.c (Module): Trying compiling on Linux with clang (instead 9: of gcc which is too slow) and lot of warnings suppressed 10: 11: 2024-06-28 Nicolas Brouard <brouard@ined.fr> 12: 13: * imach.c (Module): fixing some bugs in gnuplot and quantitative 14: variables, but not completely solved 15: 16: 2024-06-28 Nicolas Brouard <brouard@ined.fr> 17: 18: * imach.c (Module): s6 errors with age*age (harmless). 19: 20: 2024-05-12 Nicolas Brouard <brouard@ined.fr> 21: 22: * imach.c Version 0.99s5 In fact, the covariance of total life 23: expectancy e.. with a partial life expectancy e.j is high, 24: therefore the complete matrix of variance covariance has to be 25: included in the formula of the standard error of the proportion of 26: total life expectancy spent in a specific state: 27: var(X/Y)=mu_x^2/mu_y^2*(sigma_x^2/mu_x^2 -2 28: sigma_xy/mu_x/mu_y+sigma^2/mu_y^2). Also an error with mle=-3 29: made the program core dump. It is fixed in this version. 30: 31: 2024-04-30 Nicolas Brouard <brouard@ined.fr> 32: 33: * (Module): In version 0.99s4, we incorporated the calculation of 34: the std error of the proportion of total life expectancy spent in 35: a specific state Var(e.j/e..) using the formula of Var(X/Y) 36: depending only of the variances of X and Y and expectancies. 37: 38: 2024-04-24 Nicolas Brouard <brouard@ined.fr> 39: 40: * (Module): This version comes late after having tested 41: successfully the praxis C version of Buckhardt. But Buckardt's 42: version was difficult to read and Gegenfurtner's version had a few 43: typos which made its results less reliable than Buckhardt's 44: results. The most important work consisted in retyping the Brent 45: original PRAXIS program written in Algol W (published with errors, 46: ommitting the transposition of matrix V before its QR reduction 47: from Golub. I used the recent "awe" compiler from Gkynn Webster. 48: The awe library had errors, for example in arc tangent function 49: which have been fixed. 50: 51: The main objective was to get identical results with the three 52: versions: (1) Algol W, (2) Buckhardt'C version as well (3) 53: Gegenfürtner C versions on the various test functions published by 54: Brent in 1973 in Algol W. 55: 56: Also, in order to compare them, the random function had to produce 57: the same sequence for the 3 softwares. The random function used in 58: imach corresponds to original Brent's random function written in 59: Algol W. Other point, in Algol W, the arrays of dimension n are 60: 'normal' mathematical arrays starting from 1 to n. But this is a 61: real issue in C where, by default, arrays are starting from 0 to 62: n-1. In Buckhardt, as well as in Gegenfürtner C code, it can be 63: seen that authors while trying to mimick original Brent Algol W 64: code are hesitating by changing either a loop originally from 1 to 65: n in a loop from 0 to n-1, or keeping Brent's loop from 1 to n and 66: shifting the index from original X(I) in Algol W to x[i-1] in C. 67: But as IMaCh is using, since the beginning, the Numerical Recipes 68: functions vector or matrix, I changed Geggenfürtner code to mimick 69: the original Algol W arrays. Thus the X(I) is translated in C as 70: x[i] which minimizes the errors. The Golub QR algorithm was 71: published in Algol with overflow errors which were reproduced in 72: Brent's Algol W code. Buckhardt code fixed these errors which are 73: much more problematic in C than in Algol W. Thus Buckhardt code 74: seems very safe, but i haven't chosen it for IMaCh because the C 75: style is horrible and almost unreadable compared to Gegenfürtner 76: CO code which is very close to Brent's original. Also what makes 77: Buckhardt code more difficult to read is, instead of passing the 78: minimum of parameters in the functions calls, as it is in Algol 79: Brent's code or Gegenfürtner's code, the list of parameters is 80: high. For example, the flin function LONG REAL PROCEDURE FLIN 81: (LONG REAL VALUE L) has only one parameter in Algol W, the 82: Gegenfürtner flin function had two parameters: static double 83: flin(l, j) double l; { int i; double tflin[N];} but Buckhardt 84: function has 14 parameters which makes the code unreadable and 85: useless. Gegenfürtner used a lot of static variables or functions 86: which I tried to minimize. Also in Gegefürtner, array dimensions 87: were fixed to N. In my adaptation the flin is static double 88: flin(double l, int j) and the parameter used are global variables. 89: 90: 2023-06-14 Nicolas Brouard <brouard@ined.fr> 91: 92: * imach.c (Module): Testing if conjugate gradient could be quicker 93: when lot of variables POWELLORIGINCONJUGATE 94: 95: 2023-05-23 Nicolas Brouard <brouard@ined.fr> 96: 97: * imach.c (Module): Fixed PROB_r 98: 99: 2023-05-22 Nicolas Brouard <brouard@ined.fr> 100: 101: * imach.c (Module): In the ILK....txt file, the number of columns 102: before the covariates values is dependent of the number of states (16+nlstate): 0.99r46 103: 104: 2023-05-08 Nicolas Brouard <brouard@ined.fr> 105: 106: * (Module): Error V0 when result:. and model 1+age+V1: fixed 107: 108: 2023-04-29 Nicolas Brouard <brouard@ined.fr> 109: 110: * imach.c (Module): Inverting the model equation and fixingg bugs 111: in the drawings 112: "1+age+V7*V4*age+V6*V4*age+V7*V3*age+V6*V3*age+V6*V2*age+V7*age+V6*age+V4*age+V3*age+V2*age+V7*V4+V6*V4+V7*V3+V6*V3+V7*V2+V6*V2+V7+V6+V4+V3+V2\000\0003+V2" 113: 114: 2023-04-24 Nicolas Brouard <brouard@ined.fr> 115: 116: * imach.c (Module): some bugs in printinggnuplot 117: 118: 2022-09-19 Nicolas Brouard <brouard@ined.fr> 119: 120: * imach.c (Module): Version 0.99r42 needed a newer version of 121: Gnuplot. But newer version 0.99r43 should run with the Gnuplot 122: version 5.0 or 5.1 distributed with IMaCh. 123: 124: 2022-09-18 Nicolas Brouard <brouard@ined.fr> 125: 126: * imach.c (Module): Fixing when a state in the data is 0 or higher 127: than lstate+ndeath. Fixing the plot of contribution to Likelihood. 128: 129: 2022-09-16 Nicolas Brouard <brouard@ined.fr> 130: 131: * imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo 132: 133: 2022-09-14 Nicolas Brouard <brouard@ined.fr> 134: 135: * imach.c (Module): Fixing names of variables in T_ (thanks to Feinuo) 136: 137: 2022-09-14 Nicolas Brouard <brouard@ined.fr> 138: 139: * imach.c (Module): Version 0.99r39 with colored dummy covariates 140: (fixed or time varying), using new last columns of 141: ILK_parameter.txt file. 142: 143: 2022-09-11 Nicolas Brouard <brouard@ined.fr> 144: 145: * imach.c (Module): Adding timevarying products of any kinds, 146: should work before shifting cotvar from ncovcol+nqv columns in 147: order to have a correspondance between the column of cotvar and 148: the id of column. 149: (Module): Some cleaning and adding covariates in ILK.txt 150: 151: 2022-09-09 Nicolas Brouard <brouard@ined.fr> 152: 153: * imach.c (Module): Many improvements for fixing products of fixed 154: timevarying as well as fixed * fixed, and test with quantitative 155: covariate. 156: 157: 2022-09-04 Nicolas Brouard <brouard@ined.fr> 158: 159: * imach.c (Module): Now the easy runs i.e. without result or 160: model=1+age only did not work. The defautl combination should be 1 161: and not 0 because everything hasn't been tranformed yet. 162: 163: 2022-09-02 Nicolas Brouard <brouard@ined.fr> 164: 165: * imach.c: Version 0.99r35 because it outputs same results with 166: 1+age+V1+V1*age for females and 1+age for females only 167: (education=1 noweight) 168: 169: 2022-08-31 Nicolas Brouard <brouard@ined.fr> 170: 171: * imach.c (Module): Some improvments in fichtm and many verifications 0.99r34 172: 173: 2022-08-21 Nicolas Brouard <brouard@ined.fr> 174: 175: * imach.c (Module): Version 0.99r33 A lot of changes in 176: reassigning covariates: my first idea was that people will always 177: use the first covariate V1 into the model but in fact they are 178: producing data with many covariates and can use an equation model 179: with some of the covariate; it means that in a model V2+V3 instead 180: of codtabm(k,Tvaraff[j]) which calculates for combination k, for 181: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact 182: the equation model is restricted to two variables only (V2, V3) 183: and the combination for V2 should be codtabm(k,1) instead of 184: (codtabm(k,2), and the code should be 185: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been 186: made. All of these should be simplified once a day like we did in 187: hpxij() for example by using precov[nres] which is computed in 188: decoderesult for each nres of each resultline. Loop should be done 189: on the equation model globally by distinguishing only product with 190: age (which are changing with age) and no more on type of 191: covariates, single dummies, single covariates. 192: 193: 2022-08-06 Nicolas Brouard <brouard@ined.fr> 194: 195: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function 196: 197: 2022-08-03 Nicolas Brouard <brouard@ined.fr> 198: 199: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates. 200: 201: 2022-07-25 Brouard Nicolas <brouard@brouard.name> 202: 203: * imach.c (Module): Error cptcovn instead of nsd in bmij (was 204: coredumped, revealed by Feiuno, thank you. 205: 206: 2022-07-23 Nicolas Brouard <brouard@ined.fr> 207: 208: * r29 W and not sqrt(Wald) 209: 210: 2022-07-22 Nicolas Brouard <brouard@ined.fr> 211: 212: * imach.c (Module): Output of Wald test in the htm file and not only in the log. 213: 214: 2022-06-02 Brouard Nicolas <brouard@brouard.name> 215: 216: * imach.c (Module): Adding the Wald tests from the log to the main 217: htm for better display of the maximum likelihood estimators. 218: 219: 2022-05-30 Brouard Nicolas <brouard@brouard.name> 220: 221: * imach.c: With products of covariates (age or dummies or 222: quantitatives), estimates of parameters were good but when 223: estimating variances the positions of newly created covariates 224: were wrongly assigned and results were wrong. Thank to Feinua_Sun! 225: 226: 2022-05-24 Nicolas Brouard <brouard@ined.fr> 227: 228: * imach.c (Module): Some attempts to find a bug of wrong estimates 229: of confidencce intervals with product in the equation modelC 230: 231: 2022-05-15 Nicolas Brouard <brouard@ined.fr> 232: 233: * imach.c (Module): Some minor improvements 234: 235: 2022-04-13 Brouard Nicolas <brouard@brouard.name> 236: 237: * imach.c (Module): Adding link to text data files 238: 239: 2022-04-11 Nicolas Brouard <brouard@ined.fr> 240: 241: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed 242: 243: 2022-04-05 Brouard Nicolas <brouard@ined.fr> 244: 245: * imach.c (Module): Fixed covariates (dummy or quantitative) 246: with missing values have never been allowed but are ERRORS and 247: program quits. Standard deviations of fixed covariates were 248: wrongly computed. Mean and standard deviations of time varying 249: covariates are still not computed. 250: 251: 2022-03-17 Brouard Nicolas <brouard@ined.fr> 252: 253: * 99r25 Improvements in fixing discrepancies between covariates 254: missing in result line but in model. 255: 256: *Wed Mar 31 2021 Nicolas Brouard <brouard@ined.fr> 257: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett 258: 259: *2021-03-08 Nicolas Brouard <brouard@ined.fr> 260: 261: * ./ (Module): Fixed bug on result: 262: 263: *2021-02-20 Nicolas Brouard <brouard@ined.fr> 264: 265: * imach.c (Module): Fix bug on quitting after result lines! 266: (Module): Version 0.99r21 267: 268: 2021-02-12 Nicolas Brouard <brouard@ined.fr> 269: 270: * imach.c (Module): The use of a Windows BOM (huge) file is now an error 271: 272: 273: 2021-02-11 Nicolas Brouard <brouard@ined.fr> 274: 275: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed. 276: 277: 2020-02-22 Brouard Nicolas <brouard@ined.fr> 278: 279: * (Module): imach.c Update mle=-3 (for computing Life expectancy 280: and life table from the data without any state) 281: 282: 2019-05-20 Brouard Nicolas <brouard@ined.fr> 283: 284: * version.h: Summary: version 0.99.r19 285: 286: * imach.c: Summary: Projection syntax simplified 287: We can now start projections, forward or backward, from the mean date 288: of inteviews up to or down to a number of years of projection: 289: prevforecast=1 yearsfproj=15.3 mobil_average=0 290: or 291: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0 292: or 293: prevbackcast=1 yearsbproj=12.3 mobil_average=1 294: or 295: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1 296: 297: 2019-05-18 Brouard Nicolas <brouard@ined.fr> 298: 299: * imach.c: Summary: doxygen tex bug 300: 301: 2019-05-16 Brouard Nicolas <brouard@ined.fr> 302: 303: * imach.c: Summary: There was some wrong lines added 304: 305: 2019-05-09 Brouard Nicolas <brouard@ined.fr> 306: 307: * Makefile, imach.c: *** empty log message *** 308: 309: * ChangeLog: Summary: C= 310: 311: * imach.c: Summary: Some updates 312: 313: * imach.c: Summary: Before ncovmax 314: 315: * version.h: *** empty log message *** 316: 317: * imach.c: Summary: 0.99r18 unlimited number of individuals 318: 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. 319: 320: 2018-12-13 Brouard Nicolas <brouard@ined.fr> 321: 322: * imach.c: Summary: Bug for young ages (<-30) will be in r17 323: 324: 2018-12-05 Brouard Nicolas <brouard@ined.fr> 325: 326: * Makefile: Summary: 0.99r16 327: 328: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 329: 330: * imach.c: Summary: Some bugs fixed 331: 332: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 333: 334: * imach.c: Summary: Bug for young ages (<-30) will be in r17 335: 336: 2018-05-02 Brouard Nicolas <brouard@ined.fr> 337: 338: * imach.c: Summary: Some bugs fixed 339: 340: 2018-05-01 Brouard Nicolas <brouard@ined.fr> 341: 342: * imach.c: 343: Summary: Bug fixed by providing frequencies only for non missing covariates 344: 345: 2018-04-27 Brouard Nicolas <brouard@ined.fr> 346: 347: * imach.c: Summary: some minor bugs 348: 349: 2018-04-21 Brouard Nicolas <brouard@ined.fr> 350: 351: * README.txt: *** empty log message *** 352: 353: * imach.c: Summary: Some bugs fixed, valgrind tested 354: 355: 2018-04-20 Brouard Nicolas <brouard@ined.fr> 356: 357: * Makefile, README.txt, setup.iss: Summary: imach 0.99r16 358: 359: * imach.c: 360: Summary: Computing mean and stdeviation of fixed quantitative variables 361: 362: 2018-04-19 Brouard Nicolas <brouard@ined.fr> 363: 364: * imach.c: Summary: Some minor bugs fixed 365: 366: * version.h: *** empty log message *** 367: 368: 2018-02-27 Brouard Nicolas <brouard@ined.fr> 369: 370: * imach.c: *** empty log message *** 371: 372: * imach.c: Summary: Adding second argument for quitting 373: 374: 2018-02-21 Brouard Nicolas <brouard@ined.fr> 375: 376: * Makefile, README.txt, imach.c: Summary: 0.99r15 377: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c 378: 379: 2017-07-20 Brouard Nicolas <brouard@ined.fr> 380: 381: * imach.c: Summary: temporary working 382: 383: 2017-07-19 Brouard Nicolas <brouard@ined.fr> 384: 385: * imach.c: Summary: Bug for mobil_average=0 and prevforecast fixed(?) 386: 387: 2017-07-17 Brouard Nicolas <brouard@ined.fr> 388: 389: * imach.c: Summary: BOM files can be read now 390: 391: 2017-06-30 Brouard Nicolas <brouard@ined.fr> 392: 393: * imach.c: Summary: Graphs improvements 394: 395: * imach.c: Summary: Saito's color 396: 397: 2017-06-29 Brouard Nicolas <brouard@ined.fr> 398: 399: * imach.c: Summary: Version 0.99r14 400: 401: 2017-06-27 Brouard Nicolas <brouard@ined.fr> 402: 403: * imach.c: Summary: More documentation on projections 404: 405: * imach.c: Summary: Color of backprojection changed from 6 to 5(yellow) 406: 407: * imach.c: Summary: Some bug with rint 408: 409: 2017-05-24 Brouard Nicolas <brouard@ined.fr> 410: 411: * imach.c: *** empty log message *** 412: 413: 2017-05-23 Brouard Nicolas <brouard@ined.fr> 414: 415: * imach.c: Summary: Code into subroutine, cleanings 416: 417: 2017-05-18 Brouard Nicolas <brouard@ined.fr> 418: 419: * imach.c: 420: Summary: backprojection and confidence intervals of backprevalence 421: 422: 2017-05-13 Brouard Nicolas <brouard@ined.fr> 423: 424: * imach.c: Summary: temporary save for backprojection 425: 426: * imach.c: Summary: Version 0.99r13 (improvements and bugs fixed) 427: 428: 2017-04-26 Brouard Nicolas <brouard@ined.fr> 429: 430: * imach.c: Summary: imach 0.99r13 Some bugs fixed 431: 432: * imach.c: Summary: Labels in graphs 433: 434: 2017-04-24 Brouard Nicolas <brouard@ined.fr> 435: 436: * imach.c: Summary: to save 437: 438: 2017-04-18 Brouard Nicolas <brouard@ined.fr> 439: 440: * imach.c: *** empty log message *** 441: 442: 2017-04-05 Brouard Nicolas <brouard@ined.fr> 443: 444: * imach.c: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1 445: 446: 2017-04-04 Brouard Nicolas <brouard@ined.fr> 447: 448: * imach.c: Summary: Gnuplot indexations fixed (humm) 449: 450: * imach.c: 451: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3 452: 453: 2017-04-03 Brouard Nicolas <brouard@ined.fr> 454: 455: * imach.c: Summary: Version 0.99r12 456: Some cleanings, conformed with updated documentation. 457: 458: 2017-03-30 Brouard Nicolas <brouard@ined.fr> 459: 460: * version.h: *** empty log message *** 461: 462: * version.h: Summary:0.99r11 463: 464: 2017-03-29 Brouard Nicolas <brouard@ined.fr> 465: 466: * imach.c: Summary: Temp 467: 468: 2017-03-27 Brouard Nicolas <brouard@ined.fr> 469: 470: * imach.c: Summary: Temporary 471: 472: 2017-03-08 Brouard Nicolas <brouard@ined.fr> 473: 474: * imach.c: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed 475: 476: * imach.c: Summary: Fixing data parameter line 477: 478: 2016-12-15 Brouard Nicolas <brouard@ined.fr> 479: 480: * imach.c: Summary: 0.99 in progress 481: 482: 2016-09-15 Brouard Nicolas <brouard@ined.fr> 483: 484: * imach.c: *** empty log message *** 485: 486: * imach.c: Summary: not working 487: 488: 2016-09-08 Brouard Nicolas <brouard@ined.fr> 489: 490: * imach.c: Summary: continue 491: 492: 2016-09-07 Brouard Nicolas <brouard@ined.fr> 493: 494: * imach.c: Summary: Starting values from frequencies 495: 496: * imach.c: *** empty log message *** 497: 498: 2016-09-02 Brouard Nicolas <brouard@ined.fr> 499: 500: * imach.c: *** empty log message *** 501: 502: 2016-08-30 Brouard Nicolas <brouard@ined.fr> 503: 504: * imach.c: Summary: Fixing a lots 505: 506: 2016-08-29 Brouard Nicolas <brouard@ined.fr> 507: 508: * imach.c: Summary: gnuplot problem in Back projection to fix 509: 510: * imach.c: Summary: Better 511: 512: 2016-08-26 Brouard Nicolas <brouard@ined.fr> 513: 514: * imach.c: 515: Summary: Improvement in Powell output in order to copy and paste 516: 517: * imach.c: Summary: Starting tests of 0.99 518: 519: * imach.c: Summary: to valgrind 520: 521: 2016-08-25 Brouard Nicolas <brouard@ined.fr> 522: 523: * imach.c: *** empty log message *** 524: 525: 2016-08-23 Brouard Nicolas <brouard@ined.fr> 526: 527: * imach.c: *** empty log message *** 528: 529: * imach.c: Summary: not working 530: 531: 2016-08-22 Brouard Nicolas <brouard@ined.fr> 532: 533: * imach.c: Summary: not working 534: 535: * imach.c: Summary: Not working 536: 537: 2016-07-23 Brouard Nicolas <brouard@ined.fr> 538: 539: * imach.c: Summary: Completing for func too 540: 541: 2016-07-22 Brouard Nicolas <brouard@ined.fr> 542: 543: * imach.c: Summary: Fixing some arrays, still debugging 544: 545: 2016-07-21 Brouard Nicolas <brouard@ined.fr> 546: 547: * imach.c: 548: Summary: 0.99 working (more or less) for Asian Workshop on multitate methods 549: 550: 2016-07-12 Brouard Nicolas <brouard@ined.fr> 551: 552: * imach.c: Summary: temp 553: 554: * imach.c: Summary: saving but not running 555: 556: 2016-07-01 Brouard Nicolas <brouard@ined.fr> 557: 558: * imach.c: Summary: Fixes 559: 560: 2016-02-19 Brouard Nicolas <brouard@ined.fr> 561: 562: * imach-0.98r.c: Summary: Kind of 0.98r? series, starting with r7 563: 564: * setup.iss, Makefile, version.h: *** empty log message *** 565: 566: * imach.c: Summary: temporary 567: 568: 2016-02-17 Brouard Nicolas <brouard@ined.fr> 569: 570: * imach.c: Summary: Probably last 0.98 stable version 0.98r6 571: 572: 2016-02-16 Brouard Nicolas <brouard@ined.fr> 573: 574: * imach.c: Summary: minor bug 575: 576: * imach.c: Summary: 0.99r2 577: 578: 2016-02-15 Brouard Nicolas <brouard@ined.fr> 579: 580: * imach.c: *** empty log message *** 581: 582: 2016-02-12 Brouard Nicolas <brouard@ined.fr> 583: 584: * imach.c: Summary: 0.99 Back projections 585: 586: 2015-12-23 Brouard Nicolas <brouard@ined.fr> 587: 588: * imach.c: Summary: Experimental backcast 589: 590: 2015-12-18 Brouard Nicolas <brouard@ined.fr> 591: 592: * imach.c: Summary: 0.98r4 Warning and status=-2 593: 594: Version 0.98r4 is now: 595: - displaying an error when status is -1, date of interview unknown and date of death known; 596: - permitting a status -2 when the vital status is unknown at a known date of right truncation. 597: Older changes concerning s=-2, dating from 2005 have been supersed. 598: 599: * Makefile: *** empty log message *** 600: 601: 2015-12-16 Brouard Nicolas <brouard@ined.fr> 602: 603: * Makefile: *** empty log message *** 604: 605: * imach.c: Summary: 0.98r4 working 606: 607: * Makefile: Summary: linuxrpm added 608: 609: * Makefile: *** empty log message *** 610: 611: * imach.c: Summary: temporary not working 612: 613: 2015-12-11 Brouard Nicolas <brouard@ined.fr> 614: 615: * imach.c: Summary: 0.98r4 616: 617: 2015-12-08 Brouard Nicolas <brouard@ined.fr> 618: 619: * Makefile: *** empty log message *** 620: 621: 2015-11-21 Brouard Nicolas <brouard@ined.fr> 622: 623: * Makefile: 624: Summary: VBoxManage change from execute --image to run --exe and --wait-exit doesn't exist any more 625: 626: * imach.c: Summary: minor typo 627: 628: * imach.c: Summary: 0.98r3 with some graph of projected cross-sectional 629: 630: Author: Nicolas Brouard 631: 632: 2015-11-18 Brouard Nicolas <brouard@ined.fr> 633: 634: * imach.c: Summary: Start working on projected prevalences 635: 636: 2015-11-17 Brouard Nicolas <brouard@ined.fr> 637: 638: * imach.c: Summary: Adding ftolpl parameter 639: Author: N Brouard 640: 641: We had difficulties to get smoothed confidence intervals. It was due 642: to the period prevalence which wasn't computed accurately. The inner 643: parameter ftolpl is now an outer parameter of the .imach parameter 644: file after estepm. If ftolpl is small 1.e-4 and estepm too, 645: computation are long. 646: 647: * version.h: *** empty log message *** 648: 649: * imach.c: Summary: temporary 650: 651: 2015-10-27 Brouard Nicolas <brouard@ined.fr> 652: 653: * Makefile: *** empty log message *** 654: 655: * setup.iss.in: Summary: some new subdirs 656: 657: * imach.c: *** empty log message *** 658: 659: * setup.iss.in: 660: Summary: Adding gnuplot 5.1 with bin and etc for pango fonts 661: 662: 2015-10-24 Brouard Nicolas <brouard@ined.fr> 663: 664: * imach.c: *** empty log message *** 665: 666: 2015-10-23 Brouard Nicolas <brouard@ined.fr> 667: 668: * imach.c: 669: Summary: 0.98r3 some clarification for graphs on likelihood contributions 670: 671: 2015-10-01 Brouard Nicolas <brouard@ined.fr> 672: 673: * imach.c: Summary: Some new graphs of contribution to likelihood 674: 675: 2015-09-30 Brouard Nicolas <brouard@ined.fr> 676: 677: * Makefile: 678: Summary: Version read from version.h, itself read with cmake 679: 680: * imach.c: Summary: looking at better estimation of the hessian 681: 682: Also a better criteria for convergence to the period prevalence And 683: therefore adding the number of years needed to converge. (The 684: prevalence in any alive state shold sum to one 685: 686: * Makefile: *** empty log message *** 687: 688: 2015-09-22 Brouard Nicolas <brouard@ined.fr> 689: 690: * imach.c: 691: Summary: Adding some overall graph on contribution to likelihood. Might change 692: 693: 2015-09-15 Brouard Nicolas <brouard@ined.fr> 694: 695: * Makefile: *** empty log message *** 696: 697: * imach.c: Summary: 0.98r0 698: 699: - Some new graphs like survival functions 700: - Some bugs fixed like model=1+age+V2. 701: 702: * ChangeLog, Makefile: Summary: 0.98r0 703: 704: * CMakeLists.txt: Summary: Some tests for setup.iss 705: 706: 2015-09-15 Nicolas J Brouard <nbrouard@tugault.ined.fr> 707: 708: * (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc 709: 710: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local> 711: 712: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line. 713: 714: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr> 715: 716: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih) 717: 718: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local> 719: 720: * imach.c (Module): Merging 1.61 to 1.162 721: 722: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr> 723: 724: * prevlim.h (Module): 725: 726: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local> 727: 728: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested... 729: (Module): Version 0.98nR Running ok, but output format still only works for three covariates. 730: 731: 2010-04-29 brouard <brouard@ined.fr> 732: 733: * imach.c (Module): Checking covariates for more complex models 734: than V1+V2. A lot of change to be done. Unstable. 735: 736: 2010-04-26 brouard <brouard@ined.fr> 737: 738: * imach.c (Module): merging some libgsl code. Fixing computation 739: of likelione (using inter/intrapolation if mle = 0) in order to 740: get same likelihood as if mle=1. 741: Some cleaning of code and comments added. 742: 743: 2009-10-29 brouard <brouard@ined.fr> 744: 745: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code. 746: 747: 2006-06-30 Brouard Nicolas <brouard@localhost> 748: 749: * imach.c (Module): Clarifications on computing e.j 750: 751: 2006-04-28 Brouard Nicolas <brouard@localhost> 752: 753: * imach.c (Module): Yes the sum of survivors was wrong since 754: imach-114 because nhstepm was no more computed in the age 755: loop. Now we define nhstepma in the age loop. 756: (Module): In order to speed up (in case of numerous covariates) we 757: compute health expectancies (without variances) in a first step 758: and then all the health expectancies with variances or standard 759: deviation (needs data from the Hessian matrices) which slows the 760: computation. 761: In the future we should be able to stop the program is only health 762: expectancies and graph are needed without standard deviations. 763: 764: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr> 765: 766: * imach.iss (Module): 767: 768: * imach.c (Module): bugs in evsi + population file added in htm output file 769: 770: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr> 771: 772: * imach.c (Module): 773: 774: 2006-03-20 Brouard Nicolas <brouard@localhost> 775: 776: * imach.c (Module): <title> changed, corresponds to .htm file 777: name. <head> headers where missing. 778: 779: * Makefile (Module): 0.98g 780: 781: * imach.c (Module): Weights can have a decimal point as for 782: English (a comma might work with a correct LC_NUMERIC environment, 783: otherwise the weight is truncated). 784: Modification of warning when the covariates values are not 0 or 785: 1. 786: Version 0.98g 787: 788: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr> 789: 790: * imach.c (Module): Comments concerning covariates added 791: 792: * noreg-setup.iss (Module): Version 0.98f 793: 794: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup 795: 796: * imach.c (Module): refinements in the computation of lli if 797: status=-2 in order to have more reliable computation if stepm is 798: not 1 month. Version 0.98f 799: 800: 2006-03-15 Brouard Nicolas <brouard@localhost> 801: 802: * imach.c (Module): Bug if status = -2, the loglikelihood was 803: computed as likelihood omitting the logarithm. Version O.98e 804: 805: 2006-03-14 Brouard Nicolas <brouard@localhost> 806: 807: * imach.c (Module): varevsij Comments added explaining the second 808: table of variances if popbased=1 . 809: (Module): Covariances of eij, ekl added, graphs fixed, new html link. 810: (Module): Function pstamp added 811: (Module): Version 0.98d 812: 813: 2006-03-06 Brouard Nicolas <brouard@localhost> 814: 815: * imach.c (Module): Variance-covariance wrong links and 816: varian-covariance of ej. is needed (Saito). 817: 818: 2006-02-27 Brouard Nicolas <brouard@localhost> 819: 820: * imach.c (Module): One freematrix added in mlikeli! 821: 822: 2006-02-26 Brouard Nicolas <brouard@localhost> 823: 824: * imach.c (Module): Some improvements in processing parameter 825: filename with strsep. 826: 827: 2006-02-24 Brouard Nicolas <brouard@localhost> 828: 829: * imach.c (Module): Memory leaks checks with valgrind and: 830: datafile was not closed, some imatrix were not freed and on matrix 831: allocation too. 832: 833: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr> 834: 835: * strsep.c (Module): Inclusion because not in mingw32!!! 836: 837: * setup.iss.in (Module): IMACHSETUPVERSION added 838: 839: * noreg-setup.iss (Module): added to cvs tree 840: 841: * Makefile (Module): IMACHSETUPVERSION added 842: 843: 2006-01-30 BROUARD Nicolas <brouard@localhost> 844: 845: * Makefile (Module, Module): Latest Makefile 846: 847: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe 848: 849: 2006-01-27 BROUARD Nicolas <brouard@localhost> 850: 851: * noreg-setup.iss (Module): ADDED 852: 853: 2006-01-24 BROUARD Nicolas <brouard@localhost> 854: 855: * imach.c (Module): Comments (lines starting with a #) are allowed in data. 856: 857: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr> 858: 859: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg 860: 861: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X 862: 863: 2005-09-30 <lievre@PC_00595> 864: 865: * imach.c (Module): sump fixed, loop imx fixed, and simplifications. 866: (Module): If the status is missing at the last wave but we know 867: that the person is alive, then we can code his/her status as -2 868: (instead of missing=-1 in earlier versions) and his/her 869: contributions to the likelihood is 1 - Prob of dying from last 870: health status (= 1-p13= p11+p12 in the easiest case of somebody in 871: the healthy state at last known wave). Version is 0.98 872: 873: 2004-05-20 Brouard Nicolas <brouard@localhost> 874: 875: * imach.c (Repository): 876: Agnes added a direct estimation of mortality (without the need of 877: computing period prevalence and differential mortality). Thus here 878: is version 0.97a which has been distributed to some people at 879: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of 880: mortality using covariates is not done today. Estimating direct 881: mortality is a very different process because it doesn't need 882: interpolation because it is easy to get the lx from the force of 883: the mortality mux in the simplest case as for a Gompertz (log mux 884: = a + b*x . But we have been able to incorporate the new code 885: within former imach program (0.96d) without deteriorating too much 886: the understanding of the program. 887: 888: 2003-06-25 <brouard@BROUARD> 889: 890: * imach.c (Module): On windows (cygwin) function asctime_r doesn't 891: exist so I changed back to asctime which exists. 892: (Module): Version 0.96b 893: 894: 2003-06-24 <brouard@BROUARD> 895: 896: * imach.c (Module): Some bugs corrected for windows. Also, when 897: mle=-1 a template is output in file "or"mypar.txt with the design 898: of the covariance matrix to be input. 899: 900: 2003-06-25 Brouard Nicolas <brouard@homebrou> 901: 902: * imach.c (Repository): Duplicated warning errors corrected. 903: (Repository): Elapsed time after each iteration is now output. It 904: helps to forecast when convergence will be reached. Elapsed time 905: is stamped in powell. We created a new html file for the graphs 906: concerning matrix of covariance. It has extension -cov.htm. 907: 908: 2003-06-23 Brouard Nicolas <brouard@homebrou> 909: 910: * 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. 911: 912: 2003-06-18 Brouard Nicolas <brouard@homebrou> 913: 914: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96 915: 916: 2003-06-17 Brouard Nicolas <brouard@homebrou> 917: 918: 919: * imach.c (Repository): Check when date of death was earlier that 920: current date of interview. It may happen when the death was just 921: prior to the death. In this case, dh was negative and likelihood 922: was wrong (infinity). We still send an "Error" but patch by 923: assuming that the date of death was just one stepm after the 924: interview. 925: (Repository): Because some people have very long ID (first column) 926: we changed int to long in num[] and we added a new lvector for 927: memory allocation. But we also truncated to 8 characters (left 928: truncation) 929: 930: (Repository): No more line truncation errors. 931: 932: 2003-06-14 Brouard Nicolas <brouard@homebrou> 933: 934: * imach.c (Repository): Add of a routine likelione (likelihood 935: only once) which prints on a text file 936: (ilk) the contributions to the likelihood for each 937: individual/wave. 938: 939: 2003-06-13 Brouard Nicolas <brouard@homebrou> 940: * imach.c (Repository): Replace "freqsummary" at a correct 941: place. It differs from routine "prevalence" which may be called 942: many times. Probs is memory consuming and must be used with 943: parcimony. 944: 945: 2003-06-17 <brouard@BROUARD> 946: 947: * timeval.h (Module): Added included file to make use of 948: gettimeofday working on win32 with cygwin. 949: 950: 2003-05-16 Brouard Nicolas <brouard@brouard> 951: * imach.c (Module): Insert of a warning if the delay between two 952: waves is negative. 953: 954: * imach.c (Module): 955: 956: 2003-05-03 <brouard@BROUARD> 957: 958: * imach.c: Some cleaning 959: 960: 2003-03-28 <brouard@BROUARD> 961: * imach.c (Module): In version up to 0.92 likelihood was computed 962: as if date of death was unknown. Death was treated as any other 963: health state: the date of the interview describes the actual state 964: and not the date of a change in health state. The former idea was 965: to consider that at each interview the state was recorded 966: (healthy, disable or death) and IMaCh was corrected; but when we 967: introduced the exact date of death then we should have modified 968: the contribution of an exact death to the likelihood. This new 969: contribution is smaller and very dependent of the step unit 970: stepm. It is no more the probability to die between last interview 971: and month of death but the probability to survive from last 972: interview up to one month before death multiplied by the 973: probability to die within a month. Thanks to Chris 974: Jackson for correcting this bug. Former versions increased 975: mortality artificially. The bad side is that we add another loop 976: which slows down the processing. The difference can be up to 10% 977: lower mortality. 978: 979: 2002-05-30 <brouard@BROUARD> 980: 981: * imach.c (Module): Add correlation matrix of one-step 982: probabilities and covariance matrix 983: