Spatial meshing for general Bayesian multivariate models

TL;DR

This paper introduces Bayesian spatial models that handle non-Gaussian data efficiently by using directed acyclic graph-based processes and a novel SiMPA sampling algorithm, enabling scalable analysis of large geospatial datasets.

Contribution

It presents a new Bayesian modeling framework with non-Gaussian likelihoods and a specialized MCMC algorithm, improving computational efficiency for large-scale spatial data.

Findings

Demonstrates improved efficiency over existing methods in synthetic data.

Shows successful application to real-world remote sensing data.

Achieves scalable inference at hundreds of thousands of locations.

Abstract

Quantifying spatial and/or temporal associations in multivariate geolocated data of different types is achievable via spatial random effects in a Bayesian hierarchical model, but severe computational bottlenecks arise when spatial dependence is encoded as a latent Gaussian process (GP) in the increasingly common large scale data settings on which we focus. The scenario worsens in non-Gaussian models because the reduced analytical tractability leads to additional hurdles to computational efficiency. In this article, we introduce Bayesian models of spatially referenced data in which the likelihood or the latent process (or both) are not Gaussian. First, we exploit the advantages of spatial processes built via directed acyclic graphs, in which case the spatial nodes enter the Bayesian hierarchy and lead to posterior sampling via routine Markov chain Monte Carlo (MCMC) methods. Second, motivated by the possible inefficiencies of popular gradient-based sampling approaches in the multivariate contexts on which we focus, we introduce the simplified manifold preconditioner adaptation (SiMPA) algorithm which uses second order information about the target but avoids expensive matrix operations. We demostrate the performance and efficiency improvements of our methods relative to alternatives in extensive synthetic and real world remote sensing and community ecology applications with large scale data at up to hundreds of thousands of spatial locations and up to tens of outcomes. Software for the proposed methods is part of R package 'meshed', available on CRAN.