ABSTRACT
This is the main methodological chapter which discusses spatio-temporal modeling for point referenced continuous data. As in the preceding chapter, it obtains exact theoretical results for a simple separable spatio-temporal model, which serves as the baseline for model comparison purposes. More complex models, requiring advanced computation methods, are introduced gradually. Various models such as the independent GP-based model, autoregressive model, dynamic model, and the Gaussian predictive process based model are fitted and compared using the Bsptime command in the bmstdr package. The user does not need to directly code with the underlying packages such as spTimer, rstan, INLA, and spBayes. The nysptime data set is used as a running example to illustrate all the methods.
