This paper presents a Bayesian semi-parametric approach to mixed proportional hazard models. The unobserved heterogeneity and the baseline hazard rate are assumed to be an integrated Dirichlet process and an integrated weighted gamma process, respectively. Bayesian estimators and a numerical algorithm are presented.
|Translated title of the contribution
|Bayesian semi-parametric modeling for mixed proportional hazard models with right censoring
|719 - 728
|Number of pages
|Statistics and Probability Letters
|Published - Apr 2006