Return to ChangeLog CVS log

Up to [Local Repository] / imach / src

Annotation of imach/src/ChangeLog, revision 1.6

1.6     ! brouard     1: 2003-06-17  Brouard Nicolas  <brouard@homebrou>

        !             2: 

        !             3:        * imach.c (Repository): Check when date of death was earlier that

        !             4:        current date of interview. It may happen when the death was just

        !             5:        prior to the death. In this case, dh was negative and likelihood

        !             6:        was wrong (infinity). We still send an "Error" but patch by

        !             7:        assuming that the date of death was just one stepm after the

        !             8:        interview.

        !             9:        (Repository): Because some people have very long ID (first column)

        !            10:        we changed int to long in num[] and we added a new lvector for

        !            11:        memory allocation. But we also truncated to 8 characters (left

        !            12:        truncation)

        !            13:        (Repository): No more line truncation errors.

        !            14: 

        !            15: 2003-06-14  Brouard Nicolas  <brouard@homebrou>

        !            16: 

        !            17:        * imach.c (Repository): Add of a routine likelione (likelihood

        !            18:        only once) which prints on a text file

        !            19:        (ilk) the contributions to the likelihood for each

        !            20:        individual/wave.

        !            21: 

        !            22: 2003-06-13  Brouard Nicolas  <brouard@homebrou>

        !            23: 

        !            24:        * imach.c (Repository): Replace "freqsummary" at a correct

        !            25:        place. It differs from routine "prevalence" which may be called

        !            26:        many times. Probs is memory consuming and must be used with

        !            27:        parcimony.

        !            28: 

        !            29: 2003-06-17    <brouard@BROUARD>

        !            30: 

        !            31:        * timeval.h (Module): Added included file to make use of

        !            32:        gettimeofday working on win32 with cygwin.

        !            33: 

        !            34: 2003-05-16  Brouard Nicolas  <brouard@brouard>

        !            35: 

        !            36:        * imach.c (Module): Insert of a warning if the delay between two

        !            37:        waves is negative.

        !            38: 

        !            39:        * imach.c (Module): 

        !            40: 

        !            41: 2003-05-03    <brouard@BROUARD>

        !            42: 

        !            43:        * imach.c: Some cleaning

        !            44: 2003-03-28    <brouard@BROUARD>

        !            45: 

        !            46:        * imach.c (Module): In version up to 0.92 likelihood was computed

        !            47:        as if date of death was unknown. Death was treated as any other

        !            48:        health state: the date of the interview describes the actual state

        !            49:        and not the date of a change in health state. The former idea was

        !            50:        to consider that at each interview the state was recorded

        !            51:        (healthy, disable or death) and IMaCh was corrected; but when we

        !            52:        introduced the exact date of death then we should have modified

        !            53:        the contribution of an exact death to the likelihood. This new

        !            54:        contribution is smaller and very dependent of the step unit

        !            55:        stepm. It is no more the probability to die between last interview

        !            56:        and month of death but the probability to survive from last

        !            57:        interview up to one month before death multiplied by the

        !            58:        probability to die within a month. Thanks to Chris

        !            59:        Jackson for correcting this bug.  Former versions increased

        !            60:        mortality artificially. The bad side is that we add another loop

        !            61:        which slows down the processing. The difference can be up to 10%

        !            62:        lower mortality.

        !            63: 

        !            64: 2002-05-30    <brouard@BROUARD>

        !            65: 

        !            66:        * 

        !            67: 

        !            68:        * imach.c (Module): Add correlation matrix of one-step

        !            69:        probabilities and covariance matrix

        !            70: