imach/src/ChangeLog - view

File: [Local Repository] / imach / src / ChangeLog
Revision 1.72: download - view: text, annotated - select for diffs
Tue Jul 2 09:16:53 2024 UTC (3 days, 10 hours ago) by brouard
Branches: MAIN
CVS tags: HEAD

Summary: trying clang compiler on Linux

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