Bayesian Joint Models for Longitudinal and Survival Data

Abstract

This article takes a quick look at Bayesian joint models (BJMs) for longitudinal and survival data. A general formulation for BJM is examined in terms of the sampling distribution of the longitudinal and survival processes, the conditional distribution of the random effects, and the prior distribution. Next, a basic BJM defined in terms of a mixed linear model and a Cox survival regression models is discussed, and some extensions and other Bayesian topics are briefly outlined.

Bayesian survival analysis with BUGS

“…This general structure allows any type of modeling for the survival process such as frailty survival regression models, competing risks with frailties, cure models with frailties as well as linear mixed models or generalized linear mixed models for the longitudinal process. [88][89][90][91] See Armero 92 for a short review on Bayesian joint models up to date.…”