Bayesian spatio‐temporal models based on discrete convolutions
- 1 June 2008
- journal article
- research article
- Published by Wiley in Canadian Journal of Statistics
- Vol. 36 (2) , 239-258
Abstract
The authors consider a class of models for spatio‐temporal processes based on convolving independent processes with a discrete kernel that is represented by a lower triangular matrix. They study two families of models. In the first one, spatial Gaussian processes with isotropic correlations are convoluted with a kernel that provides temporal dependencies. In the second family, AR(p) processes are convoluted with a kernel providing spatial interactions. The covariance structures associated with these two families are quite rich. Their covariance functions that are stationary and separable in space and time as well as time dependent nonseparable and nonisotropic ones.Keywords
This publication has 28 references indexed in Scilit:
- Validation of climate model output using Bayesian statistical methodsClimatic Change, 2007
- Space–Time Covariance FunctionsJournal of the American Statistical Association, 2005
- Optimal Bayesian Design by Inhomogeneous Markov Chain SimulationJournal of the American Statistical Association, 2004
- Nonseparable, Stationary Covariance Functions for Space–Time DataJournal of the American Statistical Association, 2002
- Hierarchical Bayesian modeling in the environmental sciencesAllgemeines Statistisches Archiv, 2000
- Classes of Nonseparable, Spatio-Temporal Stationary Covariance FunctionsJournal of the American Statistical Association, 1999
- Priors and Component Structures in Autoregressive Time Series ModelsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1999
- DATA AUGMENTATION AND DYNAMIC LINEAR MODELSJournal of Time Series Analysis, 1994
- On Singular Wishart and Singular Multivariate Beta DistributionsThe Annals of Statistics, 1994
- ON STATIONARY PROCESSES IN THE PLANEBiometrika, 1954