ABSTRACT
This Chapter presents an overview of each stage in the practical application of model-based geostatistics to infectious disease risk stratification. A particular focus is given to understanding probabilistic map-making from the perspective of Bayesian hierarchical modelling, and hence on the role of shrinkage, hyper-parameter identification, posterior representation, and cross-validation within this framework. An introductory treatment of common data types and model structures for disease mapping is also given, along with a description of the use of high resolution covariates from satellite imaging, for the benefit of readers new to the field. Extensive referencing has been made throughout to highlight many of the excellent recent papers on this topic and to point towards new avenues of active research.
