Bayesian Dynamic Modeling for Large Multivariate Space-Time Data Sets Using Gaussian Predictive Processes

• Activity Number: 142
• Type: Invited
• Date/Time: Monday, July 30, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Andrew O. Finley*+ and Sudipto Banerjee
• Companies: Michigan State University and University of Minnesota
• Keywords: Bayesian inference ; dynamic models ; spatial processes ; predictive process ; spatial factor models

Abstract

In this talk we extend the applicability of a previously proposed class of dynamic space-time models by enabling them to accommodate large multivariate datasets. We focus on the common setting where space is viewed as continuous but time is taken to be discrete. Scalability is achieved by using a low-rank predictive process to reduce the dimensionality of the data and ease the computational burden of estimating the spatio-temporal process of interest. The proposed models are illustrated using weather station data collected over the continental United States between 1968--2011. Here our interest is to use readily available predictors, association among measurements at a given station, as well as dependence across space and time to improve prediction for incomplete station records and locations where station data does not exist.