version 1.7, 2018/04/21 21:10:20
|
version 1.9, 2022/03/15 18:40:36
|
Line 1
|
Line 1
|
Version IMaCh 0.99
|
Version IMaCh 0.99
|
Parameter files have new undocumented parameters.
|
|
Look at the archives of imach-users@listes.ined.fr for information.
|
|
|
|
Current documentation on the web site is for 0.98 (before the
|
|
introduction of varying covariates).
|
|
|
|
Interpolated Marlov Chain (IMaCh program)
|
Interpolated Marlov Chain (IMaCh program)
|
|
|
This program will be licensed under the GNU GNU GENERAL PUBLIC LICENSE.
|
This program is licensed under the GNU GNU GENERAL PUBLIC LICENSE.
|
|
|
The program estimates Health Expectancies from Cross-longitudinal
|
The program estimates Life Expectancies from Cross-longitudinal
|
studies.
|
studies as well as age-specific forward and backward prevalences in each state.
|
The main publication is Lievre, A. Brouard. N, Heathcote
|
The main publications are:
|
Ch. Estimating Health Expectancies from Cross-longitudinal surveys,
|
(1) Lièvre, A. Brouard. N, Heathcote Ch. Estimating Health
|
Mathematical Population Studies, 10(1), 2003, pp. 1-52.
|
Expectancies from Cross-longitudinal surveys, Mathematical Population Studies, 10(1), 2003, pp. 1-52.
|
|
(2) Nicolas Brouard, 2019, "Theory and applications of backward probabilities and prevalences in cross-longitudinal surveys". In: Arni S. R. Srinivasa Rao (Ed.) and C.R. Rao (Ed.), Handbook of Statistics.
|
|
Integrated population biology and modeling, Part B, Elsevier, p. 435-486.
|
|
http://hdl.handle.net/20.500.12204/AWltfxMuXMQCvuZmm9UI
|
|
|
To be brief, for each people interviewed we need the age and health
|
To be brief, for each people interviewed we need the age and
|
status at each interview. Also the month and year of death should be
|
state at each interview. Also the month and year of death should be
|
known as for birth. We model the probability to be observed in any
|
known as for birth. We model the probability to be observed in any
|
state at any age via a parametric model and estimate the parameters
|
state at any age via a parametric model and estimate the parameters
|
and their confidence interval by maximum likelihood. Many tables and
|
and their confidence interval by maximum likelihood. Many tables and
|
graphs are output in HTML. Graphs are built using gnuplot.
|
graphs are output in HTML. Graphs are built using gnuplot.
|
|
|
First versions appeared around year 2000. Currently (September 2015)
|
First versions appeared around year 2000. Currently (May 2019)
|
version 0.98q5 can run on any Unix, including OS/X and on Windows 10
|
version 0.99r19 can run on any Unix, including OS/X and on Windows 10
|
(64bit) as well as Windows XP (32bit).
|
(64bit) as well as Windows XP (32bit).
|
|
|
Binaries are usually provided for OS/X (DMG for 64bit), Linux (rpm)
|
Binaries are usually provided for OS/X (DMG for 64bit), Linux (rpm)
|
Line 33 and Windows installer (32 and 64bit).
|
Line 31 and Windows installer (32 and 64bit).
|
STRUCTURE of the binary
|
STRUCTURE of the binary
|
|
|
IMaCh is a command line program (no time for a GUI) which means that
|
IMaCh is a command line program (no time for a GUI) which means that
|
it can be run on a shell by $ imach foo.imach where foo.imach is your
|
it can be run on a shell by "$ imach foo.imach" where foo.imach is your
|
'parameter' file which is a simple text file with a few lines that
|
'parameter' file which is a simple text file with a few lines that
|
your must edit in order to describe your data as well as your model
|
your must edit in order to describe your data as well as your model
|
line. In this parameter file you give the name and path of your data
|
line. In this parameter file you give the name and path of your data
|
Line 81 efficient on Windows 64 and current bina
|
Line 79 efficient on Windows 64 and current bina
|
Intel C compiler. On OS/X the native LLVM C compiler from Apple is
|
Intel C compiler. On OS/X the native LLVM C compiler from Apple is
|
very efficient too.
|
very efficient too.
|
|
|
Usually IMaCh binaries are distributed under two subdirectories, 'bin' for the
|
Usually IMaCh binaries are distributed under two subdirectories, 'bin'
|
binaries which are imach.exe and gnuplot.exe and 'html' for some of the
|
for the binaries which are imach.exe and gnuplot.exe and 'html' for
|
documentation, and 'mytry' in order to test the program on a parameter
|
some of the documentation, and 'mytry' in order to test the program on
|
file named 'mypar.imach'.
|
a parameter file named 'mypar.imach'. In order to use IMaCh, you need
|
|
two text files, one is called the parameter file with extension .imach
|
|
or .txt which contains few lines and the data file which can be juge
|
|
according to the number of people surveyed. You can find in the
|
|
subdirectory doc of your installed package, some testing files like
|
|
biaspar.imach and mypar.imach . If you edit such a parameter file, you
|
|
will see a link to the data file, for exemple data1.txt . You can copy
|
|
both files, parameter and data file into a private directory. Then run
|
|
imach and drag the parameter file.
|
|
|
|
You can access to the main IMaCh server at
|
|
http://euroreves.ined.fr/imach but the main documentation is a wiki at
|
|
http://euroreves.ined.fr/imach/wiki . You can download the latest versions
|
|
at http://euroreves.ined.fr/imach/Download .
|
|
|
By default the file 'gnuplot.exe' is located on the same directory as
|
By default the file 'gnuplot.exe' is located on the same directory as
|
imach and is used by imach with its absolute path. Thus, if you have
|
imach and is used by imach with its absolute path. Thus, if you have
|