Bayesian evidence: can we beat MultiNest using traditional MCMC methods?

TL;DR

This paper introduces a new method to compute Bayesian evidence using standard MCMC algorithms, outperforming specialized tools like MultiNest in several cases, with applications in astrophysics and cosmology.

Contribution

A novel approach to calculate Bayesian evidence directly from standard MCMC results, eliminating the need for specialized evidence computation tools.

Findings

The new method outperforms MultiNest in multiple test cases.

It successfully computes Bayesian evidence from standard MCMC outputs.

Applicable to astrophysics, cosmology, and particle physics problems.

Abstract

Markov Chain Monte Carlo (MCMC) methods have revolutionised Bayesian data analysis over the years by making the direct computation of posterior probability densities feasible on modern workstations. However, the calculation of the prior predictive, the Bayesian evidence, has proved to be notoriously difficult with standard techniques. In this work a method is presented that lets one calculate the Bayesian evidence using nothing but the results from standard MCMC algorithms, like Metropolis-Hastings. This new method is compared to other methods like MultiNest, and greatly outperforms the latter in several cases. One of the toy problems considered in this work is the analysis of mock pulsar timing data, as encountered in pulsar timing array projects. This method is expected to be useful as well in other problems in astrophysics, cosmology and particle physics.