A Bayesian Approach for the Frailty Model using Gamma Process Prior

This paper describes a Bayesian approach to the survival model with frailty. Frailty models are increasingly popular in multivariate survival analysis since they allow us to model the association between individual survival times within clusters. The method described in this paper involves the proposal of an efficient Markov chain Monte Carlo (MCMC) algorithm with a gamma process prior for the baseline hazards function. An advantage of using gamma process priors over other approaches is that the MCMC algorithm does not require the generation of sample paths of the baseline hazard functions which is computationally demanding.