摘要：In this paper we propose a new bivariate long-term distributionbased on the Farlie-Gumbel-Morgenstern copula model. The proposedmodel allows for the presence of censored data and covariates in the cureparameter. For inferential purpose a Bayesian approach via Markov ChainMonte Carlo (MCMC) is considered. Further, some discussions on the modelselection criteria are given. In order to examine outlying and inuential observations,we develop a Bayesian case deletion inuence diagnostics basedon the Kullback-Leibler divergence. The newly developed procedures areillustrated on arti cial and real HIV data.
关键词：Bayesian approach; case deletion in;uence diagnostics; copula;modeling; long-term survival.
