Regressions in Spatially Dynamic Factor Models

This paper discusses a number of conceptual issues pertaining to the study of the relationships existing between two groups of variables which are supposed to be spatially and temporally correlated. Since it is assumed that this relationships can be studied in a reduced latent space, we provide an overview of the motivations for including spatial effects in a dynamic factor model, both from a theory-driven as well as from a data-driven perspective. Considerable attention is paid to the inferential framework necessary to carry out estimation and to the different assumptions, constraints and implications embedded in the various model specifications.