Spatial Statistics and Spatio-Temporal Data: Covariance Functions and Directional Properties

In the spatial or space-time context, specifying the correct covariance function is important to obtain efficient predictions and to understand the underlying physical process of interest.

There have been several books in recent years in the general area of spatial statistics. This book focuses on covariance and variogram functions, their role in prediction, and the proper choice of these functions in data applications. Presenting recent methods from 2004-2007 alongside more established methodology of assessing the usual assumptions on such functions such as isotropy, separability and symmetry and demonstrates the effects of incorrect covariance functions on prediction.