Return to ChangeLog CVS log

Up to [Local Repository] / imach / src

*** empty log message ***

    1: 2003-06-23  Brouard Nicolas  <brouard@homebrou>
    2: 
    3: 	* 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.
    4: 
    5: 2003-06-18  Brouard Nicolas  <brouard@homebrou>
    6: 
    7: 	* imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
    8: 
    9: 2003-06-17  Brouard Nicolas  <brouard@homebrou>
   10: 
   11: 
   12: 	* imach.c (Repository): Check when date of death was earlier that
   13: 	current date of interview. It may happen when the death was just
   14: 	prior to the death. In this case, dh was negative and likelihood
   15: 	was wrong (infinity). We still send an "Error" but patch by
   16: 	assuming that the date of death was just one stepm after the
   17: 	interview.
   18: 	(Repository): Because some people have very long ID (first column)
   19: 	we changed int to long in num[] and we added a new lvector for
   20: 	memory allocation. But we also truncated to 8 characters (left
   21: 	truncation)
   22: 
   23: 	(Repository): No more line truncation errors.
   24: 
   25: 2003-06-14  Brouard Nicolas  <brouard@homebrou>
   26: 
   27: 	* imach.c (Repository): Add of a routine likelione (likelihood
   28: 	only once) which prints on a text file
   29: 	(ilk) the contributions to the likelihood for each
   30: 	individual/wave.
   31: 
   32: 2003-06-13  Brouard Nicolas  <brouard@homebrou>
   33: 	* imach.c (Repository): Replace "freqsummary" at a correct
   34: 	place. It differs from routine "prevalence" which may be called
   35: 	many times. Probs is memory consuming and must be used with
   36: 	parcimony.
   37: 
   38: 2003-06-17    <brouard@BROUARD>
   39: 
   40: 	* timeval.h (Module): Added included file to make use of
   41: 	gettimeofday working on win32 with cygwin.
   42: 
   43: 2003-05-16  Brouard Nicolas  <brouard@brouard>
   44: 	* imach.c (Module): Insert of a warning if the delay between two
   45: 	waves is negative.
   46: 
   47: 	* imach.c (Module): 
   48: 
   49: 2003-05-03    <brouard@BROUARD>
   50: 
   51: 	* imach.c: Some cleaning
   52: 
   53: 2003-03-28    <brouard@BROUARD>
   54: 	* imach.c (Module): In version up to 0.92 likelihood was computed
   55: 	as if date of death was unknown. Death was treated as any other
   56: 	health state: the date of the interview describes the actual state
   57: 	and not the date of a change in health state. The former idea was
   58: 	to consider that at each interview the state was recorded
   59: 	(healthy, disable or death) and IMaCh was corrected; but when we
   60: 	introduced the exact date of death then we should have modified
   61: 	the contribution of an exact death to the likelihood. This new
   62: 	contribution is smaller and very dependent of the step unit
   63: 	stepm. It is no more the probability to die between last interview
   64: 	and month of death but the probability to survive from last
   65: 	interview up to one month before death multiplied by the
   66: 	probability to die within a month. Thanks to Chris
   67: 	Jackson for correcting this bug.  Former versions increased
   68: 	mortality artificially. The bad side is that we add another loop
   69: 
   70: 
   71: 	which slows down the processing. The difference can be up to 10%
   72: 
   73: 
   74: 	lower mortality.
   75: 
   76: 
   77: 
   78: 
   79: 
   80: 2002-05-30    <brouard@BROUARD>
   81: 
   82: 
   83: 
   84: 
   85: 
   86: 	* 
   87: 
   88: 
   89: 
   90: 
   91: 
   92: 	* imach.c (Module): Add correlation matrix of one-step
   93: 
   94: 
   95: 	probabilities and covariance matrix
   96: 
   97: 
   98: 
   99: 
  100: