Efficient Computation of Gaussian Likelihoods for Stationary Markov Random Field Models

TL;DR

This paper advances the efficient computation of Gaussian likelihoods for stationary Markov random field models, extending methods to irregular domains, missing data, and models with nugget effects, enabling better parameter estimation and model comparison.

Contribution

It proves a spectral convergence theorem, extends likelihood calculations to complex scenarios, and introduces a memory-efficient alternative formulation.

Findings

Exact likelihood yields superior parameter estimates.

Likelihood-based model comparison is feasible on large datasets.

Block and composite likelihood methods outperform PDE approximations.

Abstract

Rue and Held (2005) proposed a method for efficiently computing the Gaussian likelihood for stationary Markov random field models, when the data locations fall on a complete regular grid, and the model has no additive error term. The calculations rely on the availability of the covariances. We prove a theorem giving the rate of convergence of a spectral method of computing the covariances, establishing that the error decays faster than any polynomial in the size of the computing grid. We extend the exact likelihood calculations to the case of non-rectangular domains and missing values on the interior of the grid and to the case when an additive uncorrelated error term (nugget) is present in the model. We also give an alternative formulation of the likelihood that has a smaller memory burden, parts of which can be computed in parallel. We show in simulations that using the exact likelihood can give far better parameter estimates than using standard Markov random field approximations. Having access to the exact likelihood allows for model comparisons via likelihood ratios on large datasets, so as an application of the methods, we compare several state-of-the-art methods for large spatial datasets on an aerosol optical thickness dataset. We find that simple block independent likelihood and composite likelihood methods outperform stochastic partial differential equation approximations in terms of computation time and returning parameter estimates that nearly maximize the likelihood.