Comment: Microarrays, Empirical Bayes and the Two-Groups Model

TL;DR

This paper discusses the resurgence of empirical Bayes methods inspired by Efron's work, emphasizing flexible models beyond exchangeability, and highlights parallels between microarray analysis and hospital profiling for improved decision modeling.

Contribution

It advocates for more general multilevel and empirical Bayes models for random effects, moving beyond traditional exchangeable assumptions.

Findings

Encourages flexible, non-exchangeable models for empirical Bayes.

Draws parallels between microarray analysis and hospital profiling.

Suggests improved decision modeling through advanced multilevel models.

Abstract

Brad Efron's paper [arXiv:0808.0572] has inspired a return to the ideas behind Bayes, frequency and empirical Bayes. The latter preferably would not be limited to exchangeable models for the data and hyperparameters. Parallels are revealed between microarray analyses and profiling of hospitals, with advances suggesting more decision modeling for gene identification also. Then good multilevel and empirical Bayes models for random effects should be sought when regression toward the mean is anticipated.