On the computation of the marginal likelihood

TL;DR

This paper presents a method for accurately estimating the marginal likelihood of a statistical model using samples from the posterior distribution, aiding model comparison and selection.

Contribution

It introduces a consistent procedure for computing the marginal likelihood based on posterior samples, which improves upon existing methods.

Findings

Provides a reliable estimation technique for marginal likelihoods.

Facilitates model comparison through posterior sampling.

Enhances accuracy of Bayesian model evaluation.

Abstract

We describe briefly in this note a procedure for consistently estimating the marginal likelihood of a statistical model through a sample from the posterior distribution of the model parameters.