Mixture models for type II censored survival data in the presence of covariates (Q1966368)

The authors develop a Bayesian approach to the analysis of mixture models for survival data in presence of type II censoring. The normal-exponential, normal-normal and gamma-normal mixture models are concidered, and Markov chain Monte Carlo (MCMC) methods based on Gibbs sampling with Metropolis-Hastings algorithms are used to estimate the posterior quantities of interest. A numerical example illustrates the proposed methodology and some computational aspects of the MCMC algorithms.