Dear R users, A new package "msm", for multi-state modelling, is now available on CRAN. It can be used to fit continuous-time Markov models to irregularly-observed categorical processes. These models are typically used for the progression of chronic diseases, where a patient visits a doctor at irregular intervals and is diagnosed as being in one of a set of disease stages. The transition rates of the Markov process can be modelled as a function of covariates, and the models are estimated by maximum likelihood. The software can also fit a form of hidden Markov model: a multi-state model in which the stages are observed with misclassification. These might be used in cases where a disease screening test is subject to false positives or negatives. The probabilities of misclassification and the Markov chain transition rates can then be estimated simultaneously, by maximum likelihood. Covariates can be fitted to both the transition rates and the misclassification probabilities. Bug reports, reports of successes or failures, and suggestions for improvement are welcome. We would also be interested to hear what applications people might be interested in using this package for.
Christopher Jackson <chris.jackson at ic.ac.uk>, Research Associate, Department of Epidemiology and Public Health, Imperial College School of Medicine, Norfolk Place, London W2 1PG -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-announce mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-announce-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._