Silvia Bozza and Anthony O'Hagan
Dipartimento di Statistica, Universita di Venezia, Venezia, Italy and Department of Probability and Statistics, University of Sheffield, Sheffield, England
Publication details: In Between Data Science and Applied Data Analysis, M. Schader, W. Gaul and M. Vichi (eds), 165-172. Springer-Verlag. 2003.
In this paper, we address the problem of estimating the dependence structure of spatio-temporal stochastic processes. Starting from the assumption of separability, we propose a Bayesian semiparametric model that allows non-stationary spatial-temporal dependence structures. The model provides a conjoint estimation of the spatial and temporal covariance structures, with a hierarchical model internally to model the temporal dependence. A simulated case study is reported.