Workshop SDS: Registration
View Submission - SDS2022
A0184
Title: Long memory random fields on regular lattices Authors:  Luigi Ippoliti - University of Chieti Pescara (Italy) [presenting]
Angela Ferretti - University of Chieti Pescara (Italy)
Rajendra Bhansali - Imperial College London (United Kingdom)
Pasquale Valentini - University of Chieti-Pescara (Italy)
Abstract: The motivation is drawn from applications in environmental sciences where empirical evidence of slow decay of correlations has been found for data observed on a regular lattice. Different types of models have been proposed for analysing such data. We introduce the class of conditional autoregressive fractional integrated moving average (CARFIMA) models as a suitable framework for studying environmental data with long-range spatial correlation. For this class, we provide detailed descriptions of some representative models, make the necessary comparison with some other existing models, and discuss some important inferential and computational issues on estimation, simulation and long memory process approximation. Results from model fit comparison and predictive performance of CARFIMA models are discussed through a statistical analysis of satellite land surface temperature data.