Subgroup Analysis from Bayesian Perspectives

Abstract

Identifying the sub-population structures along with the tailored treatments for all groups plays a critical rule for assigning the best available treatment to an individual patient. Subgroup analysis, a key to develop personalized medicine, becomes increasingly important over the past decade. Besides frequentist methods, there are a spectrum of methods developed from Bayesian perspectives to identify subgroups. In this chapter, we provide a comprehensive overview of Bayesian methods and discuss their properties. We further examine empirical performance of the two Bayesian methods via simulation studies and a real data analysis.