Abstract:Multivariate spatially-oriented data sets are prevalent in the environmental and physical sciences. Scientists seek to jointly model multiple variables, each indexed by a spatial location, to capture any underlying spatial association for each variable and associations among the different dependent variables. Multivariate latent spatial process models have proved effective in driving statistical inference and rendering better predictive inference at arbitrary locations for the spatial process. High-dimensional multivariate spatial data, which is the theme of this article, refers to data sets where the number of spatial locations and the number of spatially dependent variables is very large. The field has witnessed substantial developments in scalable models for univariate spatial processes, but such methods for multivariate spatial processes, especially when the number of outcomes are moderately large, are limited in comparison. Here, we extend scalable modeling strategies for a single process to multivariate processes. We pursue Bayesian inference which is attractive for full uncertainty quantification of the latent spatial process. Our approach exploits distribution theory for the Matrix-Normal distribution, which we use to construct scalable versions of a hierarchical linear model of coregionalization (LMC) and spatial factor models that deliver inference over a high-dimensional parameter space including the latent spatial process. We illustrate the computational and inferential benefits of our algorithms over competing methods using simulation studies and an analysis of a massive vegetation index data set.

Bayesian Predictive Analysis of the Multiple Equation Linear Model System

Bayesian estimation for longitudinal data in a joint model with HPCs

Bayesian Analysis of Nonlinear Reproductive Dispersion Mixed Models

The Superiority of Bayes Estimators in A Multivariate Linear Model with Respect to Normal-Inverse Wishart Prior

Bayesian Analysis of Joint Mean and Covariance Models for Longitudinal Data

Bayesian Prediction With Linear Dynamic Model: Principle And Application

Longitudinal Data Analysis Based on Bayesian Semiparametric Method

Spatial Factor Modeling: A Bayesian Matrix-Normal Approach for Misaligned Data

Bayesian Model for Factor Analysis Using Mixing Prior Distribution

Bayesian Analysis of Non-Linear Structural Equation Models with Non-Ignorable Missing Outcomes from Reproductive Dispersion Models.

Multiple Model Estimation Represented by Bayesian Networks

Bayesian estimation of variance factor in the nonlinear model

Bayesian analysis for finite mixture in non-recursive non-linear structural equation models.

Semiparametric Bayesian Analysis of Transformation Linear Mixed Models

Bayesian analysis of linear regression models with autoregressive symmetrical errors and incomplete data

Bayesian inference for a covariance matrix

Bayesian Inference for Multiplicative Seasonal Models with Stable Innovations

Bayesian Approach to Linear Bayesian Networks

Bayesian Ying Yang System, Best Harmony Learning, And Gaussian Manifold Based Family

Bayesian Analysis for Over-parameterized Linear Model without Sparsity

Bayesian Inference and Prediction Analysis of the Power Law Process Based on A Natural Conjugate Prior