Return to ChangeLog CVS log

Up to [Local Repository] / imach / src

Annotation of imach/src/ChangeLog, revision 1.4

1.3       brouard     1: 2003-05-16  Brouard Nicolas  <brouard@brouard>
                      2: 
1.4     ! brouard     3:        * imach.c (Module): Insert of a warning if the delay between two
        !             4:        waves is negative.
        !             5: 
1.3       brouard     6:        * imach.c (Module): 
                      7: 
1.2       brouard     8: 2003-05-03    <brouard@BROUARD>
                      9: 
                     10:        * imach.c: Some cleaning
1.1       brouard    11: 2003-03-28    <brouard@BROUARD>
                     12: 
                     13:        * imach.c (Module): In version up to 0.92 likelihood was computed
                     14:        as if date of death was unknown. Death was treated as any other
                     15:        health state: the date of the interview describes the actual state
                     16:        and not the date of a change in health state. The former idea was
                     17:        to consider that at each interview the state was recorded
                     18:        (healthy, disable or death) and IMaCh was corrected; but when we
                     19:        introduced the exact date of death then we should have modified
                     20:        the contribution of an exact death to the likelihood. This new
                     21:        contribution is smaller and very dependent of the step unit
                     22:        stepm. It is no more the probability to die between last interview
                     23:        and month of death but the probability to survive from last
                     24:        interview up to one month before death multiplied by the
                     25:        probability to die within a month. Thanks to Chris
                     26:        Jackson for correcting this bug.  Former versions increased
                     27:        mortality artificially. The bad side is that we add another loop
                     28:        which slows down the processing. The difference can be up to 10%
                     29:        lower mortality.
                     30: 
                     31: 2002-05-30    <brouard@BROUARD>
                     32: 
                     33:        * 
                     34: 
                     35:        * imach.c (Module): Add correlation matrix of one-step
                     36:        probabilities and covariance matrix
                     37: