Using Weibull Mixture Distributions to Model Heterogeneous Survival Data

ABSTRACT

In this article we use Bayesian methods to fit a Weibull mixture model with an unknown number of components to possibly right-censored survival data. This is done using the recently developed, birth-death MCMC algorithm. We also show how to estimate the survivor function and the expected hazard rate from the MCMC output.

Keywords:

• Bayesian
• MCMC
• Mixtures
• Survival analysis
• Weibull

Mathematics Subject Classification:

• 62F15
• 62N99