CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Simple spatio-temporal models for complex spatial data Authors:  Lara Fontanella - University of Chieti-Pescara (Italy)
Rosaria Ignaccolo - University of Turin (Italy) [presenting]
Luigi Ippoliti - University of Chieti Pescara (Italy)
Pasquale Valentini - University of Chieti-Pescara (Italy)
Abstract: The focus is on the specification of a simple hierarchical generalized spatio-temporal model which warrants consideration when data sets with different types of spatial complexities are available. The model is a three-level hierarchical one, with a component specified by means of drift functions (e.g. we use a set of principal kriging functions or principal splines). Especially under Gaussian assumptions, the model is simple to estimate and particularly useful when reliable estimates of the parameters of a covariance function are difficult to obtain. Results from the analysis of different datasets have shown that our model can provide accurate predictions.