Introduction to Spatio-Temporal Statistics
Christopher K. Wikle, Andrew Zammit‐Mangion, Noel Cressie · 2019
From a statistical perspective, dynamic models are closer to the kinds of statistical models studied in time series than to those studied in spatial statistics. Yet, there are two fundamental differences between spatio-temporal statistical models that are dynamic, and the usual multivariate time-series models. Spatio-temporal interactions may then become apparent, and the first steps of spatio-temporal modeling can be taken. Spatio-temporal interactions are not limited to similar types of processes nor to spatial and temporal scales of variability that seem obvious. The descriptive approach typically seeks to characterize the spatio-temporal process in terms of its mean function and its covariance function. The descriptive approach often relies on an important statistical characteristic of dependent data, namely that nearby observations tend to be more alike than those far apart. The chapter also presents an overview of the key concepts discussed in this book.