Spatial Matérn Fields Driven by Non-Gaussian Noise

• Activity Number: 6
• Type: Invited
• Date/Time: Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): David Bolin*+
• Companies: Lund University
• Keywords: non-Gaussian ; Random fields ; Matérn covariances

Abstract

In this work, we study non-Gaussian extensions of a recently discovered link between certain Gaussian random fields, expressed as solutions to stochastic partial differential equations (SPDEs), and Gaussian Markov random fields. We show how to construct efficient representations of non-Gaussian random fields driven by Generalized asymmetric Laplace (GAL) noise and Normal inverse Gaussian (NIG) noise, and discuss how to do parameter estimation and spatial prediction for these models. Finally, we look at an application to precipitation data from the US where we compare the results obtained using our non-Gaussian latent models with results obtained using standard Gaussian models for transformed data.