A HIERARCHICAL BAYESIAN APPROACH TO THE ESTIMATION OF MONOTONE HAZARD RATES IN THE RANDOM RIGHT CENSORSHIP MODEL

Abstract

Here we study hierarchical Bayesian estimation of a monotone hazard rate for
both complete and randomly right censored data. We propose two methods
of computation: Monte-Carlo importance sampling and Laplace approximation
techniques. These methods are computationally simple and easily implemented
on complex hazard functions. They are compared in simulation studies with uncensored
and censored data and the methodology is illustrated on two interesting
data sets.