imach/src/ChangeLog - diff

Return to ChangeLog CVS log

Up to [Local Repository] / imach / src

Diff for /imach/src/ChangeLog between versions 1.63 and 1.72

-version 1.63, 2023/04/29 10:53:23
+version 1.72, 2024/07/02 09:16:53
  Line 1
- <<<<<<< ChangeLog
+-07-02  Nicolas Brouard  <brouard@ined.fr>
+         * imach.c (Module): Trying compiling on Linux with clang (instead
+         of gcc which is too slow) and lot of warnings suppressed
+-06-28  Nicolas Brouard  <brouard@ined.fr>
+         * imach.c (Module): s6 errors with age*age (harmless).
+-05-12  Nicolas Brouard   <brouard@ined.fr>
+         * imach.c Version 0.99s5 In fact, the covariance of total life
+         expectancy e.. with a partial life expectancy e.j is high,
+         therefore the complete matrix of variance covariance has to be
+         included in the formula of the standard error of the proportion of
+         total life expectancy spent in a specific state:
+         var(X/Y)=mu_x^2/mu_y^2*(sigma_x^2/mu_x^2 -2
+         sigma_xy/mu_x/mu_y+sigma^2/mu_y^2).  Also an error with mle=-3
+         made the program core dump. It is fixed in this version.
+-04-30  Nicolas Brouard   <brouard@ined.fr>
+         * (Module): In version 0.99s4, we incorporated the calculation of
+         the std error of the proportion of total life expectancy spent in
+         a specific state Var(e.j/e..) using the formula of Var(X/Y)
+         depending only of the variances of X and Y and expectancies.
+-04-24  Nicolas Brouard   <brouard@ined.fr>
+         * (Module): This version comes late after having tested
+         successfully the praxis C version of Buckhardt.  But Buckardt's
+         version was difficult to read and Gegenfurtner's version had a few
+         typos which made its results less reliable than Buckhardt's
+         results.  The most important work consisted in retyping the Brent
+         original PRAXIS program written in Algol W (published with errors,
+         ommitting the transposition of matrix V before its QR reduction
+         from Golub. I used the recent "awe" compiler from Gkynn Webster.
+         The awe library had errors, for example in arc tangent function
+         which have been fixed.
+         The main objective was to get identical results with the three
+         versions: (1) Algol W, (2) Buckhardt'C version as well (3)
+         Gegenfürtner C versions on the various test functions published by
+         Brent in 1973 in Algol W.
+         Also, in order to compare them, the random function had to produce
+         the same sequence for the 3 softwares. The random function used in
+         imach corresponds to original Brent's random function written in
+         Algol W.  Other point, in Algol W, the arrays of dimension n are
+         'normal' mathematical arrays starting from 1 to n. But this is a
+         real issue in C where, by default, arrays are starting from 0 to
+         n-1. In Buckhardt, as well as in Gegenfürtner C code, it can be
+         seen that authors while trying to mimick original Brent Algol W
+         code are hesitating by changing either a loop originally from 1 to
+         n in a loop from 0 to n-1, or keeping Brent's loop from 1 to n and
+         shifting the index from original X(I) in Algol W to x[i-1] in C.
+         But as IMaCh is using, since the beginning, the Numerical Recipes
+         functions vector or matrix, I changed Geggenfürtner code to mimick
+         the original Algol W arrays. Thus the X(I) is translated in C as
+         x[i] which minimizes the errors. The Golub QR algorithm was
+         published in Algol with overflow errors which were reproduced in
+         Brent's Algol W code. Buckhardt code fixed these errors which are
+         much more problematic in C than in Algol W.  Thus Buckhardt code
+         seems very safe, but i haven't chosen it for IMaCh because the C
+         style is horrible and almost unreadable compared to Gegenfürtner
+         CO code which is very close to Brent's original. Also what makes
+         Buckhardt code more difficult to read is, instead of passing the
+         minimum of parameters in the functions calls, as it is in Algol
+         Brent's code or Gegenfürtner's code, the list of parameters is
+         high. For example, the flin function LONG REAL PROCEDURE FLIN
+         (LONG REAL VALUE L) has only one parameter in Algol W, the
+         Gegenfürtner flin function had two parameters: static double
+         flin(l, j) double l; { int i; double tflin[N];} but Buckhardt
+         function has 14 parameters which makes the code unreadable and
+         useless. Gegenfürtner used a lot of static variables or functions
+         which I tried to minimize. Also in Gegefürtner, array dimensions
+         were fixed to N. In my adaptation the flin is static double
+         flin(double l, int j) and the parameter used are global variables.
+-06-14  Nicolas Brouard   <brouard@ined.fr>
+         * imach.c (Module): Testing if conjugate gradient could be quicker
+         when lot of variables POWELLORIGINCONJUGATE
+-05-23  Nicolas Brouard   <brouard@ined.fr>
+         * imach.c (Module): Fixed PROB_r
+-05-22  Nicolas Brouard   <brouard@ined.fr>
+         * imach.c (Module): In the ILK....txt file, the number of columns
+         before the covariates values is dependent of the number of states (16+nlstate): 0.99r46
+-05-08  Nicolas Brouard   <brouard@ined.fr>
+         *  (Module): Error V0 when result:. and model 1+age+V1: fixed
 -04-29  Nicolas Brouard  <brouard@ined.fr>
          * imach.c (Module): Inverting the model equation and fixingg bugs
- Line 108
+ Line 204
 -06-02  Brouard Nicolas  <brouard@brouard.name>
          * imach.c (Module): Adding the Wald tests from the log to the main
          htm for better display of the maximum likelihood estimators.
 -05-30  Brouard Nicolas  <brouard@brouard.name>
          * imach.c: With products of covariates (age or dummies or
          quantitatives), estimates of parameters were good but when
          estimating variances the positions of newly created covariates
          were wrongly assigned and results were wrong. Thank to Feinua_Sun!
 -05-24  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module): Some attempts to find a bug of wrong estimates
          of confidencce intervals with product in the equation modelC
 -05-15  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module):  Some minor improvements
 -04-13  Brouard Nicolas  <brouard@brouard.name>
          * imach.c (Module): Adding link to text data files
 -04-11  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
 -04-05  Brouard Nicolas  <brouard@ined.fr>
          * imach.c (Module): Fixed covariates (dummy or quantitative)
          with missing values have never been allowed but are ERRORS and
          program quits. Standard deviations of fixed covariates were
          wrongly computed. Mean and standard deviations of time varying
          covariates are still not computed.
 -03-17  Brouard Nicolas  <brouard@ined.fr>
          * 99r25 Improvements in fixing discrepancies between covariates
          missing in result line but in model.
  *Wed Mar 31 2021  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
  *2021-03-08  Nicolas Brouard   <brouard@ined.fr>
          * ./ (Module): Fixed bug on result:
  *2021-02-20  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module): Fix bug on quitting after result lines!
          (Module): Version 0.99r21
 -02-12  Nicolas Brouard   <brouard@ined.fr>
          * imach.c (Module): The use of a Windows BOM (huge) file is now an error
 -02-11  Nicolas  Brouard   <brouard@ined.fr>
          *  (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
 -02-22  Brouard Nicolas  <brouard@ined.fr>
          *  (Module): imach.c Update mle=-3 (for computing Life expectancy
          and life table from the data without any state)
 -05-20  Brouard Nicolas  <brouard@ined.fr>
          * version.h: Summary: version 0.99.r19
          * imach.c: Summary: Projection syntax simplified
          We can now start projections, forward or backward, from the mean date
          of inteviews up to or down to a number of years of projection:
          prevforecast=1 yearsfproj=15.3 mobil_average=0
          or
          prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
          or
          prevbackcast=1 yearsbproj=12.3 mobil_average=1
          or
          prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
 -05-18  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: doxygen tex bug
 -05-16  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: There was some wrong lines added
 -05-09  Brouard Nicolas  <brouard@ined.fr>
          * Makefile, imach.c: *** empty log message ***
          * ChangeLog: Summary: C=
          * imach.c: Summary: Some updates
          * imach.c: Summary: Before ncovmax
          * version.h: *** empty log message ***
          * imach.c: Summary: 0.99r18 unlimited number of individuals
          The number n which was limited to 20,000 cases is now unlimited, from firstobs to lastobs. If the number is too for the virtual memory, probably an error will occur.
 -12-13  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Bug for young ages (<-30) will be in r17
 -12-05  Brouard Nicolas  <brouard@ined.fr>
          * Makefile: Summary: 0.99r16
 -05-02  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Some bugs fixed
 -05-02  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:  Summary: Bug for young ages (<-30) will be in r17
 -05-02  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Some bugs fixed
 -05-01  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: Bug fixed by providing frequencies only for non missing covariates
 -04-27  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: some minor bugs
 -04-21  Brouard Nicolas  <brouard@ined.fr>
          * README.txt: *** empty log message ***
          * imach.c: Summary: Some bugs fixed, valgrind tested
 -04-20  Brouard Nicolas  <brouard@ined.fr>
          * Makefile, README.txt, setup.iss: Summary: imach 0.99r16
          * imach.c:
          Summary: Computing mean and stdeviation of fixed quantitative variables
 -04-19  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Some minor bugs fixed
          * version.h: *** empty log message ***
 -02-27  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
          * imach.c: Summary: Adding second argument for quitting
 -02-21  Brouard Nicolas  <brouard@ined.fr>
          * Makefile, README.txt, imach.c: Summary: 0.99r15
          New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
 -07-20  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: temporary working
 -07-19  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
 -07-17  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: BOM files can be read now
 -06-30  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Graphs improvements
          * imach.c: Summary: Saito's color
 -06-29  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Version 0.99r14
 -06-27  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: More documentation on projections
          * imach.c: Summary: Color of backprojection changed from 6 to 5(yellow)
          * imach.c: Summary: Some bug with rint
 -05-24  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -05-23  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Code into subroutine, cleanings
 -05-18  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: backprojection and confidence intervals of backprevalence
 -05-13  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: temporary save for backprojection
          * imach.c: Summary: Version 0.99r13 (improvements and bugs fixed)
 -04-26  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: imach 0.99r13 Some bugs fixed
          * imach.c: Summary: Labels in graphs
 -04-24  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: to save
 -04-18  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -04-05  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
 -04-04  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Gnuplot indexations fixed (humm)
          * imach.c:
          Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
 -04-03  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Version 0.99r12
          Some cleanings, conformed with updated documentation.
 -03-30  Brouard Nicolas  <brouard@ined.fr>
          * version.h: *** empty log message ***
          * version.h: Summary:0.99r11
 -03-29  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Temp
 -03-27  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Temporary
 -03-08  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
          * imach.c: Summary: Fixing data parameter line
 -12-15  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: 0.99 in progress
 -09-15  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
          * imach.c: Summary: not working
 -09-08  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: continue
 -09-07  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Starting values from frequencies
          * imach.c: *** empty log message ***
 -09-02  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -08-30  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Fixing a lots
 -08-29  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: gnuplot problem in Back projection to fix
          * imach.c: Summary: Better
 -08-26  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: Improvement in Powell output in order to copy and paste
          * imach.c: Summary: Starting tests of 0.99
          * imach.c: Summary: to valgrind
 -08-25  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -08-23  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
          * imach.c: Summary: not working
 -08-22  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: not working
          * imach.c: Summary: Not working
 -07-23  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Completing for func too
 -07-22  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Fixing some arrays, still debugging
 -07-21  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: 0.99 working (more or less) for Asian Workshop on multitate methods
 -07-12  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: temp
          * imach.c: Summary: saving but not running
 -07-01  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Fixes
 -02-19  Brouard Nicolas  <brouard@ined.fr>
          * imach-0.98r.c: Summary: Kind of 0.98r? series, starting with r7
          * setup.iss, Makefile, version.h: *** empty log message ***
          * imach.c: Summary: temporary
 -02-17  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Probably last 0.98 stable version 0.98r6
 -02-16  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: minor bug
          * imach.c: Summary: 0.99r2
 -02-15  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -02-12  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: 0.99 Back projections
 -12-23  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Experimental backcast
 -12-18  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: 0.98r4 Warning and status=-2
          Version 0.98r4 is now:
           - displaying an error when status is -1, date of interview unknown and date of death known;
           - permitting a status -2 when the vital status is unknown at a known date of right truncation.
          Older changes concerning s=-2, dating from 2005 have been supersed.
          * Makefile: *** empty log message ***
 -12-16  Brouard Nicolas  <brouard@ined.fr>
          * Makefile: *** empty log message ***
          * imach.c: Summary: 0.98r4 working
          * Makefile: Summary: linuxrpm added
          * Makefile: *** empty log message ***
          * imach.c: Summary: temporary not working
 -12-11  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: 0.98r4
 -12-08  Brouard Nicolas  <brouard@ined.fr>
          * Makefile: *** empty log message ***
 -11-21  Brouard Nicolas  <brouard@ined.fr>
          * Makefile:
          Summary: VBoxManage change from execute --image to run --exe and --wait-exit doesn't exist any more
          * imach.c: Summary: minor typo
          * imach.c: Summary: 0.98r3 with some graph of projected cross-sectional
          Author: Nicolas Brouard
 -11-18  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Start working on projected prevalences
 -11-17  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Adding ftolpl parameter
          Author: N Brouard
          We had difficulties to get smoothed confidence intervals. It was due
          to the period prevalence which wasn't computed accurately. The inner
          parameter ftolpl is now an outer parameter of the .imach parameter
          file after estepm. If ftolpl is small 1.e-4 and estepm too,
          computation are long.
          * version.h: *** empty log message ***
          * imach.c: Summary: temporary
 -10-27  Brouard Nicolas  <brouard@ined.fr>
          * Makefile: *** empty log message ***
          * setup.iss.in: Summary: some new subdirs
          * imach.c: *** empty log message ***
          * setup.iss.in:
          Summary: Adding gnuplot 5.1 with bin and etc for pango fonts
 -10-24  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: *** empty log message ***
 -10-23  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: 0.98r3 some clarification for graphs on likelihood contributions
 -10-01  Brouard Nicolas  <brouard@ined.fr>
          * imach.c: Summary: Some new graphs of contribution to likelihood
 -09-30  Brouard Nicolas  <brouard@ined.fr>
          * Makefile:
          Summary: Version read from version.h, itself read with cmake
          * imach.c: Summary: looking at better estimation of the hessian
          Also a better criteria for convergence to the period prevalence And
          therefore adding the number of years needed to converge. (The
          prevalence in any alive state shold sum to one
          * Makefile: *** empty log message ***
 -09-22  Brouard Nicolas  <brouard@ined.fr>
          * imach.c:
          Summary: Adding some overall graph on contribution to likelihood. Might change
 -09-15  Brouard Nicolas  <brouard@ined.fr>
          * Makefile: *** empty log message ***
          * imach.c: Summary: 0.98r0
          - Some new graphs like survival functions
          - Some bugs fixed like model=1+age+V2.
          * ChangeLog, Makefile: Summary: 0.98r0
          * CMakeLists.txt: Summary: Some tests for setup.iss
 -09-15  Nicolas J Brouard   <nbrouard@tugault.ined.fr>
          *  (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc
 -08-18  Nicolas J Brouard   <nbrouard@tugault.local>
          *  (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
 -08-03  Nicolas J Brouard   <nbrouard@tugault.ined.fr>
          *  (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih)
 -12-16  Nicolas J Brouard   <nbrouard@tugault.local>
          * imach.c (Module): Merging 1.61 to 1.162
 -09-02  Nicolas J Brouard   <nbrouard@njbrouard.ined.fr>
          * prevlim.h (Module):
 -01-26  Nicolas J Brouard   <nbrouard@ledermann-2.local>
          * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
          (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
 -04-29  brouard  <brouard@ined.fr>
          * imach.c (Module): Checking covariates for more complex models
          than V1+V2. A lot of change to be done. Unstable.
 -04-26  brouard  <brouard@ined.fr>
          * imach.c (Module): merging some libgsl code. Fixing computation
          of likelione (using inter/intrapolation if mle = 0) in order to
          get same likelihood as if mle=1.
          Some cleaning of code and comments added.
 -10-29  brouard  <brouard@ined.fr>
          * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
 -06-30  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Clarifications on computing e.j
 -04-28  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Yes the sum of survivors was wrong since
          imach-114 because nhstepm was no more computed in the age
          loop. Now we define nhstepma in the age loop.
          (Module): In order to speed up (in case of numerous covariates) we
          compute health expectancies (without variances) in a first step
          and then all the health expectancies with variances or standard
          deviation (needs data from the Hessian matrices) which slows the
          computation.
          In the future we should be able to stop the program is only health
          expectancies and graph are needed without standard deviations.
 -04-04  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
          * imach.iss (Module):
          * imach.c (Module): bugs in evsi + population file added in htm output file
 -03-22  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
          * imach.c (Module):
 -03-20  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): <title> changed, corresponds to .htm file
          name. <head> headers where missing.
          * Makefile (Module): 0.98g
          * imach.c (Module): Weights can have a decimal point as for
          English (a comma might work with a correct LC_NUMERIC environment,
          otherwise the weight is truncated).
          Modification of warning when the covariates values are not 0 or
 .
          Version 0.98g
 -03-16  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
          * imach.c (Module): Comments concerning covariates added
          * noreg-setup.iss (Module): Version 0.98f
          * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
          * imach.c (Module): refinements in the computation of lli if
          status=-2 in order to have more reliable computation if stepm is
          not 1 month. Version 0.98f
 -03-15  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Bug if status = -2, the loglikelihood was
          computed as likelihood omitting the logarithm. Version O.98e
 -03-14  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): varevsij Comments added explaining the second
          table of variances if popbased=1 .
          (Module): Covariances of eij, ekl added, graphs fixed, new html link.
          (Module): Function pstamp added
          (Module): Version 0.98d
 -03-06  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Variance-covariance wrong links and
          varian-covariance of ej. is needed (Saito).
 -02-27  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): One freematrix added in mlikeli!
 -02-26  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Some improvements in processing parameter
          filename with strsep.
 -02-24  Brouard Nicolas  <brouard@localhost>
          * imach.c (Module): Memory leaks checks with valgrind and:
          datafile was not closed, some imatrix were not freed and on matrix
          allocation too.
 -02-09  LIEVRE Agnes  <lievre@PC_00697.ined.fr>
          * strsep.c (Module): Inclusion because not in mingw32!!!
          * setup.iss.in (Module): IMACHSETUPVERSION added
          * noreg-setup.iss (Module): added to cvs tree
          * Makefile (Module): IMACHSETUPVERSION added
 -01-30  BROUARD Nicolas  <brouard@localhost>
          * Makefile (Module, Module): Latest Makefile
          * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
 -01-27  BROUARD Nicolas  <brouard@localhost>
          * noreg-setup.iss (Module): ADDED
 -01-24  BROUARD Nicolas  <brouard@localhost>
          * imach.c (Module): Comments (lines starting with a #) are allowed in data.
 -10-25  Nicolas Brouard  <brouard@dhcp171.recherche.ined.fr>
          * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
          * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
 -09-30    <lievre@PC_00595>
          * 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
          contributions to the likelihood is 1 - Prob of dying from last
          health status (= 1-p13= p11+p12 in the easiest case of somebody in
          the healthy state at last known wave). Version is 0.98
 -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.
 -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
 -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.
 -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.
 -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.
 -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
 -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.
 -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.
 -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.
 -06-17    <brouard@BROUARD>
          * timeval.h (Module): Added included file to make use of
          gettimeofday working on win32 with cygwin.
 -05-16  Brouard Nicolas  <brouard@brouard>
          * imach.c (Module): Insert of a warning if the delay between two
          waves is negative.
          * imach.c (Module):
 -05-03    <brouard@BROUARD>
          * imach.c: Some cleaning
 -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.
 -05-30    <brouard@BROUARD>
          * imach.c (Module): Add correlation matrix of one-step
          probabilities and covariance matrix

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