
This book, published in the Springer Series in Statistics, is an expository account of the modelbased aproach to geostatistical analysis. In contrast to classical geostatistics, the modelbased approach sets geostatistical methods in the context of generic statistical modelling, with the likelihood playing a central role in inference, whether Bayesian or frequentist. Throughout the book, methods are illustrated with applications to data from epidemiology and the environmental sciences. Almost all of the analyses reported in the book can be reproduced by readers using the authors’ geoR software.

