Generalized spatio-temporal models

Abstract

An important problem in statistics is the study of spatio-te mporal data taking into account the effect of explanatory variables such as latitude, longitud e and time. In this paper, a new Bayesian approach for analyzing spatial longitudinal data is propos ed. It takes into account linear time regression structures on the mean and linear regression str uctures on the variance-covariance matrix of normal observations. The spatial structure is inc luded in the time regression parameters and also in the regression structure of the variance covaria nce matrix. Initially, we present a summary of the spatial models and the Bayesian methodology u sed to fit the models, as a extension of the longitudinal data analysis. Next, the gene ral spatial temporal model is proposed. Finally, this proposal is used to study rainfall data