A major update of the "msm" package for continuous-time Markov and hidden Markov multi-state models is now available from CRAN. Hidden Markov models with general, continuous response distributions are now supported. These models are used for Markov processes which can only be observed through the value of some noisy marker. Censored states are now supported at any observation time. The observation scheme is also allowed to be different for each observation, for example, process snapshot / fully-observed histories / partially-observed histories such as death times. Users upgrading from previous versions of msm should note that some of the syntax for specifying models has changed. In particular, initial values are now specified in the "qmatrix" and "covinits" arguments instead of the inits argument, so that "qmatrix" is no longer a matrix of 0/1 indicators. As always, refer to the user guide "msm-manual.pdf" in the "doc" subdirectory for a full description of the methodology, and worked examples in the use of the package. See the appendix for a full list of the syntax changes, and the NEWS file in the top-level installation directory for a full list of the changes in this release.
Christopher Jackson <chris.jackson at imperial.ac.uk>, Research Associate, Department of Epidemiology and Public Health, Imperial College School of Medicine, Norfolk Place, London W2 1PG, tel. 020 759 43371 _______________________________________________ R-packages mailing list R-packages at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-packages