ABSTRACT
Much of this text has focused upon spatial process models that presume, for a region of study D, a collection of random variables {Y (s) : s ∈ D}, which can be viewed as a randomly realized surface over the region. In practice, this surface is only observed at a finite set of locations and inferential interest typically resides in estimation of the process parameters as well as in spatial interpolation of the process at the unobserved locations.
