Markov Chain Monte Carlo with the Integrated Nested Laplace Approximation

TL;DR

This paper introduces a novel method combining INLA and MCMC to extend Bayesian hierarchical model fitting capabilities beyond traditional Gaussian Markov random field assumptions, enabling more flexible model analysis.

Contribution

The paper presents a new approach that integrates INLA with MCMC to allow fitting of a broader class of models, including those with parameters that are fixed or conditional.

Findings

Extends INLA applicability to more complex models.

Demonstrates the method with simple and advanced examples.

Shows how to use R-INLA with MCMC for model fitting.

Abstract

The Integrated Nested Laplace Approximation (INLA) has established itself as a widely used method for approximate inference on Bayesian hierarchical models which can be represented as a latent Gaussian model (LGM). INLA is based on producing an accurate approximation to the posterior marginal distributions of the parameters in the model and some other quantities of interest by using repeated approximations to intermediate distributions and integrals that appear in the computation of the posterior marginals. INLA focuses on models whose latent effects are a Gaussian Markov random field (GMRF). For this reason, we have explored alternative ways of expanding the number of possible models that can be fitted using the INLA methodology. In this paper, we present a novel approach that combines INLA and Markov chain Monte Carlo (MCMC). The aim is to consider a wider range of models that cannot be fitted with INLA unless some of the parameters of the model have been fixed. Hence, conditioning on these parameters the model could be fitted with the R-INLA package. We show how new values of these parameters can be drawn from their posterior by using conditional models fitted with INLA and standard MCMC algorithms, such as Metropolis-Hastings. Hence, this will extend the use of INLA to fit models that can be expressed as a conditional LGM. Also, this new approach can be used to build simpler MCMC samplers for complex models as it allows sampling only on a limited number parameters in the model. We will demonstrate how our approach can extend the class of models that could benefit from INLA, and how the R-INLA package will ease its implementation. We will go through simple examples of this new approach before we discuss more advanced problems with datasets taken from relevant literature.