Bayesian model averaging with the integrated nested Laplace approximation

TL;DR

This paper discusses the use of Bayesian model averaging with the integrated nested Laplace approximation (INLA) to efficiently estimate posterior distributions in Bayesian hierarchical models, especially spatial econometrics models.

Contribution

It reviews the application of BMA with INLA and introduces a new example involving spatial econometrics models.

Findings

BMA with INLA improves model fitting efficiency.

The approach enables handling multiple models in spatial econometrics.

INLA provides fast posterior marginal estimates.

Abstract

The integrated nested Laplace approximation (INLA) for Bayesian inference is an efficient approach to estimate the posterior marginal distributions of the parameters and latent effects of Bayesian hierarchical models that can be expressed as latent Gaussian Markov random fields (GMRF). The representation as a GMRF allows the associated software R-INLA to estimate the posterior marginals in a fraction of the time as typical Markov chain Monte Carlo algorithms. INLA can be extended by means of Bayesian model averaging (BMA) to increase the number of models that it can fit to conditional latent GMRF. In this paper we review the use of BMA with INLA and propose a new example on spatial econometrics models.