|
Readme file
SERIES C
 Applied Statistics
Models for longitudinal data with censored changepoints, by
C.H.Jackson and L.D. Sharples
Journal of the Royal Statistical Society,
Series C, Applied Statistics, Volume
53 (2003) part 1, 149-162
Data sets used in the analysis
fev.dat
Columns correspond to patient ID, months after transplant, percentage
of baseline FEV1 and an indicator for the occurrence of an acute event
within 14 days of the observation time
earlyacute.dat
Columns correspond to patient ID, a dummy variable of all ones, the
number of acute rejections in the first six months after transplant,
an indicator for the occurrence of CMV in the first six months, and
the number of episodes of other types of infection within the first
six months.
Program source code files
The program is written in C++ and consists of the following source
files: data.cc, main.cc, options.cc, output.cc, params.cc, routines.cc,
arms.c, random.c, matrix.c, mix.hh, routines.hh, arms.h, random.h, matrix.h
The file 'Makefile' is used to control compilation of the sources.
The files medium.cmd, medium.in and priors.in contain options for specifying
the model to be run, the location of the data and optionsfor controlling
the MCMC.
The program has only been tested on Linux and Solaris systems, using
the C++ compiler g++ (part of the GNU compiler collection "gcc").
Instructions for use of the program
Brief instructions for how to construct the configuration files and
run the program are given in the file MANUAL.txt.
Christopher Jackson
Department of Epidemiology and Public Health
Imperial College School of Medicine
Norfolk Place
London
W2 1PG
UK
E-mail: chris.jackson@imperial.ac.uk
|
Journals SERIES
A
Statistics
in Society
SERIES B
Statistical
Methodology
SERIES C
Applied
Statistics
SERIES D
The
Statistician
|
