File:  [Local Repository] / imach / src / ChangeLog
Revision 1.32: download - view: text, annotated - select for diffs
Thu Oct 29 13:18:35 2009 UTC (14 years, 7 months ago) by brouard
Branches: MAIN
CVS tags: HEAD
*** empty log message ***

    1: 2009-10-29  brouard  <brouard@ined.fr>
    2: 
    3: 	* imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
    4: 
    5: 2006-06-30  Brouard Nicolas  <brouard@localhost>
    6: 
    7: 	* imach.c (Module): Clarifications on computing e.j
    8: 
    9: 2006-04-28  Brouard Nicolas  <brouard@localhost>
   10: 
   11: 	* imach.c (Module): Yes the sum of survivors was wrong since
   12: 	imach-114 because nhstepm was no more computed in the age
   13: 	loop. Now we define nhstepma in the age loop.
   14: 	(Module): In order to speed up (in case of numerous covariates) we
   15: 	compute health expectancies (without variances) in a first step
   16: 	and then all the health expectancies with variances or standard
   17: 	deviation (needs data from the Hessian matrices) which slows the
   18: 	computation.
   19: 	In the future we should be able to stop the program is only health
   20: 	expectancies and graph are needed without standard deviations.
   21: 
   22: 2006-04-04  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
   23: 
   24: 	* imach.iss (Module): 
   25: 
   26: 	* imach.c (Module): bugs in evsi + population file added in htm output file
   27: 
   28: 2006-03-22  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
   29: 
   30: 	* imach.c (Module): 
   31: 
   32: 2006-03-20  Brouard Nicolas  <brouard@localhost>
   33: 
   34: 	* imach.c (Module): <title> changed, corresponds to .htm file
   35: 	name. <head> headers where missing.
   36: 
   37: 	* Makefile (Module): 0.98g
   38: 
   39: 	* imach.c (Module): Weights can have a decimal point as for
   40: 	English (a comma might work with a correct LC_NUMERIC environment,
   41: 	otherwise the weight is truncated).
   42: 	Modification of warning when the covariates values are not 0 or
   43: 	1. 
   44: 	Version 0.98g
   45: 
   46: 2006-03-16  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
   47: 
   48: 	* imach.c (Module): Comments concerning covariates added
   49: 
   50: 	* noreg-setup.iss (Module): Version 0.98f
   51: 
   52: 	* Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
   53: 
   54: 	* imach.c (Module): refinements in the computation of lli if
   55: 	status=-2 in order to have more reliable computation if stepm is
   56: 	not 1 month. Version 0.98f
   57: 
   58: 2006-03-15  Brouard Nicolas  <brouard@localhost>
   59: 
   60: 	* imach.c (Module): Bug if status = -2, the loglikelihood was
   61: 	computed as likelihood omitting the logarithm. Version O.98e
   62: 
   63: 2006-03-14  Brouard Nicolas  <brouard@localhost>
   64: 
   65: 	* imach.c (Module): varevsij Comments added explaining the second
   66: 	table of variances if popbased=1 .
   67: 	(Module): Covariances of eij, ekl added, graphs fixed, new html link.
   68: 	(Module): Function pstamp added
   69: 	(Module): Version 0.98d 
   70: 
   71: 2006-03-06  Brouard Nicolas  <brouard@localhost>
   72: 
   73: 	* imach.c (Module): Variance-covariance wrong links and
   74: 	varian-covariance of ej. is needed (Saito).
   75: 
   76: 2006-02-27  Brouard Nicolas  <brouard@localhost>
   77: 
   78: 	* imach.c (Module): One freematrix added in mlikeli!
   79: 
   80: 2006-02-26  Brouard Nicolas  <brouard@localhost>
   81: 
   82: 	* imach.c (Module): Some improvements in processing parameter
   83: 	filename with strsep.
   84: 
   85: 2006-02-24  Brouard Nicolas  <brouard@localhost>
   86: 
   87: 	* imach.c (Module): Memory leaks checks with valgrind and:
   88: 	datafile was not closed, some imatrix were not freed and on matrix
   89: 	allocation too.
   90: 
   91: 2006-02-09  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
   92: 
   93: 	* strsep.c (Module): Inclusion because not in mingw32!!!
   94: 
   95: 	* setup.iss.in (Module): IMACHSETUPVERSION added
   96: 
   97: 	* noreg-setup.iss (Module): added to cvs tree
   98: 
   99: 	* Makefile (Module): IMACHSETUPVERSION added
  100: 
  101: 2006-01-30  BROUARD Nicolas  <brouard@localhost>
  102: 
  103: 	* Makefile (Module, Module): Latest Makefile
  104: 
  105: 	* imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
  106: 
  107: 2006-01-27  BROUARD Nicolas  <brouard@localhost>
  108: 
  109: 	* noreg-setup.iss (Module): ADDED
  110: 
  111: 2006-01-24  BROUARD Nicolas  <brouard@localhost>
  112: 
  113: 	* imach.c (Module): Comments (lines starting with a #) are allowed in data.
  114: 
  115: 2005-10-25  Nicolas Brouard  <brouard@dhcp171.recherche.ined.fr>
  116: 
  117: 	* Makefile (Module): Added Imach98.pmsp and howto crete a dmg
  118: 
  119: 	* createdmg.sh (Module): Creates a dmg disk for MaC OS/X
  120: 
  121: 2005-09-30    <lievre@PC_00595>
  122: 
  123: 	* imach.c (Module): sump fixed, loop imx fixed, and simplifications.
  124: 	(Module): If the status is missing at the last wave but we know
  125: 	that the person is alive, then we can code his/her status as -2
  126: 	(instead of missing=-1 in earlier versions) and his/her
  127: 	contributions to the likelihood is 1 - Prob of dying from last
  128: 	health status (= 1-p13= p11+p12 in the easiest case of somebody in
  129: 	the healthy state at last known wave). Version is 0.98 
  130: 
  131: 2004-05-20  Brouard Nicolas  <brouard@localhost>
  132: 
  133: 	* imach.c (Repository): 
  134: 	Agnes added a direct estimation of mortality (without the need of
  135: 	computing period prevalence and differential mortality). Thus here
  136: 	is version 0.97a which has been distributed to some people at
  137: 	REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
  138: 	mortality using covariates is not done today. Estimating direct
  139: 	mortality is a very different process because it doesn't need
  140: 	interpolation because it is easy to get the lx from the force of
  141: 	the mortality mux in the simplest case as for a Gompertz (log mux
  142: 	= a + b*x . But we have been able to incorporate the new code
  143: 	within former imach program (0.96d) without deteriorating too much
  144: 	the understanding of the program. 
  145: 
  146: 2003-06-25    <brouard@BROUARD>
  147: 
  148: 	* imach.c (Module): On windows (cygwin) function asctime_r doesn't
  149: 	exist so I changed back to asctime which exists.
  150: 	(Module): Version 0.96b
  151: 
  152: 2003-06-24    <brouard@BROUARD>
  153: 
  154: 	* imach.c (Module): Some bugs corrected for windows. Also, when
  155: 	mle=-1 a template is output in file "or"mypar.txt with the design
  156: 	of the covariance matrix to be input.
  157: 
  158: 2003-06-25  Brouard Nicolas  <brouard@homebrou>
  159: 
  160: 	* imach.c (Repository): Duplicated warning errors corrected.
  161: 	(Repository): Elapsed time after each iteration is now output. It
  162: 	helps to forecast when convergence will be reached. Elapsed time
  163: 	is stamped in powell.  We created a new html file for the graphs
  164: 	concerning matrix of covariance. It has extension -cov.htm.
  165: 
  166: 2003-06-23  Brouard Nicolas  <brouard@homebrou>
  167: 
  168: 	* 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.
  169: 
  170: 2003-06-18  Brouard Nicolas  <brouard@homebrou>
  171: 
  172: 	* imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
  173: 
  174: 2003-06-17  Brouard Nicolas  <brouard@homebrou>
  175: 
  176: 
  177: 	* imach.c (Repository): Check when date of death was earlier that
  178: 	current date of interview. It may happen when the death was just
  179: 	prior to the death. In this case, dh was negative and likelihood
  180: 	was wrong (infinity). We still send an "Error" but patch by
  181: 	assuming that the date of death was just one stepm after the
  182: 	interview.
  183: 	(Repository): Because some people have very long ID (first column)
  184: 	we changed int to long in num[] and we added a new lvector for
  185: 	memory allocation. But we also truncated to 8 characters (left
  186: 	truncation)
  187: 
  188: 	(Repository): No more line truncation errors.
  189: 
  190: 2003-06-14  Brouard Nicolas  <brouard@homebrou>
  191: 
  192: 	* imach.c (Repository): Add of a routine likelione (likelihood
  193: 	only once) which prints on a text file
  194: 	(ilk) the contributions to the likelihood for each
  195: 	individual/wave.
  196: 
  197: 2003-06-13  Brouard Nicolas  <brouard@homebrou>
  198: 	* imach.c (Repository): Replace "freqsummary" at a correct
  199: 	place. It differs from routine "prevalence" which may be called
  200: 	many times. Probs is memory consuming and must be used with
  201: 	parcimony.
  202: 
  203: 2003-06-17    <brouard@BROUARD>
  204: 
  205: 	* timeval.h (Module): Added included file to make use of
  206: 	gettimeofday working on win32 with cygwin.
  207: 
  208: 2003-05-16  Brouard Nicolas  <brouard@brouard>
  209: 	* imach.c (Module): Insert of a warning if the delay between two
  210: 	waves is negative.
  211: 
  212: 	* imach.c (Module): 
  213: 
  214: 2003-05-03    <brouard@BROUARD>
  215: 
  216: 	* imach.c: Some cleaning
  217: 
  218: 2003-03-28    <brouard@BROUARD>
  219: 	* imach.c (Module): In version up to 0.92 likelihood was computed
  220: 	as if date of death was unknown. Death was treated as any other
  221: 	health state: the date of the interview describes the actual state
  222: 	and not the date of a change in health state. The former idea was
  223: 	to consider that at each interview the state was recorded
  224: 	(healthy, disable or death) and IMaCh was corrected; but when we
  225: 	introduced the exact date of death then we should have modified
  226: 	the contribution of an exact death to the likelihood. This new
  227: 	contribution is smaller and very dependent of the step unit
  228: 	stepm. It is no more the probability to die between last interview
  229: 	and month of death but the probability to survive from last
  230: 	interview up to one month before death multiplied by the
  231: 	probability to die within a month. Thanks to Chris
  232: 	Jackson for correcting this bug.  Former versions increased
  233: 	mortality artificially. The bad side is that we add another loop
  234: 
  235: 
  236: 	which slows down the processing. The difference can be up to 10%
  237: 
  238: 
  239: 	lower mortality.
  240: 
  241: 
  242: 
  243: 
  244: 
  245: 2002-05-30    <brouard@BROUARD>
  246: 
  247: 
  248: 
  249: 
  250: 
  251: 	* 
  252: 
  253: 
  254: 
  255: 
  256: 
  257: 	* imach.c (Module): Add correlation matrix of one-step
  258: 
  259: 
  260: 	probabilities and covariance matrix
  261: 
  262: 
  263: 
  264: 
  265: 

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>