ABSTRACT
In this final chapter, the models and methodology presented in previous chapters for continuous-scale univariate spatial data in Euclidean domains are extended to five additional types of spatial data: global data (data extending over much of the surface of the earth), data on stream networks, space-time data, multivariate spatial data, and discrete (binary or count) spatial data. In particular, for each type of data we present methods for characterizing and modeling the spatial (and spatio-temporal or multivariate) dependence, for estimating the effects of explanatory variables on the response variable(s), and for predicting the variable(s) at unsampled locations and times. Unique features of these extensions are highlighted, but the treatment of topics is comparatively superficial. Many of the topics covered are currently active areas of research.
