Diff for /imach/src/ChangeLog between versions 1.6 and 1.22

version 1.6, 2003/06/17 20:16:05 version 1.22, 2006/03/15 17:43:07
Line 1 Line 1
 2003-06-17  Brouard Nicolas  <brouard@homebrou>  2006-03-15  Brouard Nicolas  <brouard@localhost>
   
         * imach.c (Repository): Check when date of death was earlier that          * imach.c (Module): Bug if status = -2, the loglikelihood was
         current date of interview. It may happen when the death was just          computed as likelihood omitting the logarithm. Version O.98e
         prior to the death. In this case, dh was negative and likelihood  
         was wrong (infinity). We still send an "Error" but patch by  2006-03-14  Brouard Nicolas  <brouard@localhost>
         assuming that the date of death was just one stepm after the  
         interview.          * imach.c (Module): varevsij Comments added explaining the second
         (Repository): Because some people have very long ID (first column)          table of variances if popbased=1 .
         we changed int to long in num[] and we added a new lvector for          (Module): Covariances of eij, ekl added, graphs fixed, new html link.
         memory allocation. But we also truncated to 8 characters (left          (Module): Function pstamp added
         truncation)          (Module): Version 0.98d
         (Repository): No more line truncation errors.  
   2006-03-06  Brouard Nicolas  <brouard@localhost>
 2003-06-14  Brouard Nicolas  <brouard@homebrou>  
           * imach.c (Module): Variance-covariance wrong links and
         * imach.c (Repository): Add of a routine likelione (likelihood          varian-covariance of ej. is needed (Saito).
         only once) which prints on a text file  
         (ilk) the contributions to the likelihood for each  2006-02-27  Brouard Nicolas  <brouard@localhost>
         individual/wave.  
           * imach.c (Module): One freematrix added in mlikeli!
 2003-06-13  Brouard Nicolas  <brouard@homebrou>  
   2006-02-26  Brouard Nicolas  <brouard@localhost>
         * imach.c (Repository): Replace "freqsummary" at a correct  
         place. It differs from routine "prevalence" which may be called          * imach.c (Module): Some improvements in processing parameter
         many times. Probs is memory consuming and must be used with          filename with strsep.
         parcimony.  
   2006-02-24  Brouard Nicolas  <brouard@localhost>
 2003-06-17    <brouard@BROUARD>  
           * imach.c (Module): Memory leaks checks with valgrind and:
         * timeval.h (Module): Added included file to make use of          datafile was not closed, some imatrix were not freed and on matrix
         gettimeofday working on win32 with cygwin.          allocation too.
   
 2003-05-16  Brouard Nicolas  <brouard@brouard>  2006-02-09  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
   
         * imach.c (Module): Insert of a warning if the delay between two          * strsep.c (Module): Inclusion because not in mingw32!!!
         waves is negative.  
           * setup.iss.in (Module): IMACHSETUPVERSION added
         * imach.c (Module):  
           * noreg-setup.iss (Module): added to cvs tree
 2003-05-03    <brouard@BROUARD>  
           * Makefile (Module): IMACHSETUPVERSION added
         * imach.c: Some cleaning  
 2003-03-28    <brouard@BROUARD>  2006-01-30  BROUARD Nicolas  <brouard@localhost>
   
         * imach.c (Module): In version up to 0.92 likelihood was computed          * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
         as if date of death was unknown. Death was treated as any other  
         health state: the date of the interview describes the actual state  2006-01-27  BROUARD Nicolas  <brouard@localhost>
         and not the date of a change in health state. The former idea was  
         to consider that at each interview the state was recorded          * noreg-setup.iss (Module): ADDED
         (healthy, disable or death) and IMaCh was corrected; but when we  
         introduced the exact date of death then we should have modified  2006-01-24  BROUARD Nicolas  <brouard@localhost>
         the contribution of an exact death to the likelihood. This new  
         contribution is smaller and very dependent of the step unit          * imach.c (Module): Comments (lines starting with a #) are allowed in data.
         stepm. It is no more the probability to die between last interview  
         and month of death but the probability to survive from last  2005-10-25  Nicolas Brouard  <brouard@dhcp171.recherche.ined.fr>
         interview up to one month before death multiplied by the  
         probability to die within a month. Thanks to Chris          * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
         Jackson for correcting this bug.  Former versions increased  
         mortality artificially. The bad side is that we add another loop          * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
         which slows down the processing. The difference can be up to 10%  
         lower mortality.  2005-09-30    <lievre@PC_00595>
   
 2002-05-30    <brouard@BROUARD>          * imach.c (Module): sump fixed, loop imx fixed, and simplifications.
           (Module): If the status is missing at the last wave but we know
         *          that the person is alive, then we can code his/her status as -2
           (instead of missing=-1 in earlier versions) and his/her
         * imach.c (Module): Add correlation matrix of one-step          contributions to the likelihood is 1 - Prob of dying from last
         probabilities and covariance matrix          health status (= 1-p13= p11+p12 in the easiest case of somebody in
           the healthy state at last known wave). Version is 0.98
   
   2004-05-20  Brouard Nicolas  <brouard@localhost>
   
           * imach.c (Repository):
           Agnes added a direct estimation of mortality (without the need of
           computing period prevalence and differential mortality). Thus here
           is version 0.97a which has been distributed to some people at
           REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
           mortality using covariates is not done today. Estimating direct
           mortality is a very different process because it doesn't need
           interpolation because it is easy to get the lx from the force of
           the mortality mux in the simplest case as for a Gompertz (log mux
           = a + b*x . But we have been able to incorporate the new code
           within former imach program (0.96d) without deteriorating too much
           the understanding of the program.
   
   2003-06-25    <brouard@BROUARD>
   
           * imach.c (Module): On windows (cygwin) function asctime_r doesn't
           exist so I changed back to asctime which exists.
           (Module): Version 0.96b
   
   2003-06-24    <brouard@BROUARD>
   
           * imach.c (Module): Some bugs corrected for windows. Also, when
           mle=-1 a template is output in file "or"mypar.txt with the design
           of the covariance matrix to be input.
   
   2003-06-25  Brouard Nicolas  <brouard@homebrou>
   
           * imach.c (Repository): Duplicated warning errors corrected.
           (Repository): Elapsed time after each iteration is now output. It
           helps to forecast when convergence will be reached. Elapsed time
           is stamped in powell.  We created a new html file for the graphs
           concerning matrix of covariance. It has extension -cov.htm.
   
   2003-06-23  Brouard Nicolas  <brouard@homebrou>
   
           * 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.
   
   2003-06-18  Brouard Nicolas  <brouard@homebrou>
   
           * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
   
   2003-06-17  Brouard Nicolas  <brouard@homebrou>
   
   
           * imach.c (Repository): Check when date of death was earlier that
           current date of interview. It may happen when the death was just
           prior to the death. In this case, dh was negative and likelihood
           was wrong (infinity). We still send an "Error" but patch by
           assuming that the date of death was just one stepm after the
           interview.
           (Repository): Because some people have very long ID (first column)
           we changed int to long in num[] and we added a new lvector for
           memory allocation. But we also truncated to 8 characters (left
           truncation)
   
           (Repository): No more line truncation errors.
   
   2003-06-14  Brouard Nicolas  <brouard@homebrou>
   
           * imach.c (Repository): Add of a routine likelione (likelihood
           only once) which prints on a text file
           (ilk) the contributions to the likelihood for each
           individual/wave.
   
   2003-06-13  Brouard Nicolas  <brouard@homebrou>
           * imach.c (Repository): Replace "freqsummary" at a correct
           place. It differs from routine "prevalence" which may be called
           many times. Probs is memory consuming and must be used with
           parcimony.
   
   2003-06-17    <brouard@BROUARD>
   
           * timeval.h (Module): Added included file to make use of
           gettimeofday working on win32 with cygwin.
   
   2003-05-16  Brouard Nicolas  <brouard@brouard>
           * imach.c (Module): Insert of a warning if the delay between two
           waves is negative.
   
           * imach.c (Module):
   
   2003-05-03    <brouard@BROUARD>
   
           * imach.c: Some cleaning
   
   2003-03-28    <brouard@BROUARD>
           * imach.c (Module): In version up to 0.92 likelihood was computed
           as if date of death was unknown. Death was treated as any other
           health state: the date of the interview describes the actual state
           and not the date of a change in health state. The former idea was
           to consider that at each interview the state was recorded
           (healthy, disable or death) and IMaCh was corrected; but when we
           introduced the exact date of death then we should have modified
           the contribution of an exact death to the likelihood. This new
           contribution is smaller and very dependent of the step unit
           stepm. It is no more the probability to die between last interview
           and month of death but the probability to survive from last
           interview up to one month before death multiplied by the
           probability to die within a month. Thanks to Chris
           Jackson for correcting this bug.  Former versions increased
           mortality artificially. The bad side is that we add another loop
   
   
           which slows down the processing. The difference can be up to 10%
   
   
           lower mortality.
   
   
   
   
   
   2002-05-30    <brouard@BROUARD>
   
   
   
   
   
           *
   
   
   
   
   
           * imach.c (Module): Add correlation matrix of one-step
   
   
           probabilities and covariance matrix
   
   
   
   
   

Removed from v.1.6  
changed lines
  Added in v.1.22


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