Return to ChangeLog CVS log

Up to [Local Repository] / imach / src

*** empty log message ***

    1: 2004-05-20  Brouard Nicolas  <brouard@localhost>
    2: 
    3: 	* imach.c (Repository): 
    4: 	Agnes added a direct estimation of mortality (without the need of
    5: 	computing period prevalence and differential mortality). Thus here
    6: 	is version 0.97a which has been distributed to some people at
    7: 	REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
    8: 	mortality using covariates is not done today. Estimating direct
    9: 	mortality is a very different process because it doesn't need
   10: 	interpolation because it is easy to get the lx from the force of
   11: 	the mortality mux in the simplest case as for a Gompertz (log mux
   12: 	= a + b*x . But we have been able to incorporate the new code
   13: 	within former imach program (0.96d) without deteriorating too much
   14: 	the understanding of the program. 
   15: 
   16: 2003-06-25    <brouard@BROUARD>
   17: 
   18: 	* imach.c (Module): On windows (cygwin) function asctime_r doesn't
   19: 	exist so I changed back to asctime which exists.
   20: 	(Module): Version 0.96b
   21: 
   22: 2003-06-24    <brouard@BROUARD>
   23: 
   24: 	* imach.c (Module): Some bugs corrected for windows. Also, when
   25: 	mle=-1 a template is output in file "or"mypar.txt with the design
   26: 	of the covariance matrix to be input.
   27: 
   28: 2003-06-25  Brouard Nicolas  <brouard@homebrou>
   29: 
   30: 	* imach.c (Repository): Duplicated warning errors corrected.
   31: 	(Repository): Elapsed time after each iteration is now output. It
   32: 	helps to forecast when convergence will be reached. Elapsed time
   33: 	is stamped in powell.  We created a new html file for the graphs
   34: 	concerning matrix of covariance. It has extension -cov.htm.
   35: 
   36: 2003-06-23  Brouard Nicolas  <brouard@homebrou>
   37: 
   38: 	* 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.
   39: 
   40: 2003-06-18  Brouard Nicolas  <brouard@homebrou>
   41: 
   42: 	* imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
   43: 
   44: 2003-06-17  Brouard Nicolas  <brouard@homebrou>
   45: 
   46: 
   47: 	* imach.c (Repository): Check when date of death was earlier that
   48: 	current date of interview. It may happen when the death was just
   49: 	prior to the death. In this case, dh was negative and likelihood
   50: 	was wrong (infinity). We still send an "Error" but patch by
   51: 	assuming that the date of death was just one stepm after the
   52: 	interview.
   53: 	(Repository): Because some people have very long ID (first column)
   54: 	we changed int to long in num[] and we added a new lvector for
   55: 	memory allocation. But we also truncated to 8 characters (left
   56: 	truncation)
   57: 
   58: 	(Repository): No more line truncation errors.
   59: 
   60: 2003-06-14  Brouard Nicolas  <brouard@homebrou>
   61: 
   62: 	* imach.c (Repository): Add of a routine likelione (likelihood
   63: 	only once) which prints on a text file
   64: 	(ilk) the contributions to the likelihood for each
   65: 	individual/wave.
   66: 
   67: 2003-06-13  Brouard Nicolas  <brouard@homebrou>
   68: 	* imach.c (Repository): Replace "freqsummary" at a correct
   69: 	place. It differs from routine "prevalence" which may be called
   70: 	many times. Probs is memory consuming and must be used with
   71: 	parcimony.
   72: 
   73: 2003-06-17    <brouard@BROUARD>
   74: 
   75: 	* timeval.h (Module): Added included file to make use of
   76: 	gettimeofday working on win32 with cygwin.
   77: 
   78: 2003-05-16  Brouard Nicolas  <brouard@brouard>
   79: 	* imach.c (Module): Insert of a warning if the delay between two
   80: 	waves is negative.
   81: 
   82: 	* imach.c (Module): 
   83: 
   84: 2003-05-03    <brouard@BROUARD>
   85: 
   86: 	* imach.c: Some cleaning
   87: 
   88: 2003-03-28    <brouard@BROUARD>
   89: 	* imach.c (Module): In version up to 0.92 likelihood was computed
   90: 	as if date of death was unknown. Death was treated as any other
   91: 	health state: the date of the interview describes the actual state
   92: 	and not the date of a change in health state. The former idea was
   93: 	to consider that at each interview the state was recorded
   94: 	(healthy, disable or death) and IMaCh was corrected; but when we
   95: 	introduced the exact date of death then we should have modified
   96: 	the contribution of an exact death to the likelihood. This new
   97: 	contribution is smaller and very dependent of the step unit
   98: 	stepm. It is no more the probability to die between last interview
   99: 	and month of death but the probability to survive from last
  100: 	interview up to one month before death multiplied by the
  101: 	probability to die within a month. Thanks to Chris
  102: 	Jackson for correcting this bug.  Former versions increased
  103: 	mortality artificially. The bad side is that we add another loop
  104: 
  105: 
  106: 	which slows down the processing. The difference can be up to 10%
  107: 
  108: 
  109: 	lower mortality.
  110: 
  111: 
  112: 
  113: 
  114: 
  115: 2002-05-30    <brouard@BROUARD>
  116: 
  117: 
  118: 
  119: 
  120: 
  121: 	* 
  122: 
  123: 
  124: 
  125: 
  126: 
  127: 	* imach.c (Module): Add correlation matrix of one-step
  128: 
  129: 
  130: 	probabilities and covariance matrix
  131: 
  132: 
  133: 
  134: 
  135: