Computational methods for Bayesian model choice

TL;DR

This paper surveys recent computational techniques for approximating the Bayes factor in Bayesian model selection, including importance sampling, harmonic mean sampling, and nested sampling, providing a unified perspective.

Contribution

It offers a unified review of recent methods for Bayesian model choice, clarifying their relationships and comparative advantages.

Findings

Reassessment of importance sampling, harmonic mean sampling, and nested sampling.

Comparison of methods from a unified perspective.

Insights into the strengths and limitations of each approach.

Abstract

In this note, we shortly survey some recent approaches on the approximation of the Bayes factor used in Bayesian hypothesis testing and in Bayesian model choice. In particular, we reassess importance sampling, harmonic mean sampling, and nested sampling from a unified perspective.