B0361
Title: Generalized spatial matrix specifications
Authors: Samantha Leorato - University of Milan (Italy) [presenting]
Andrea Martinelli - University of Insubria (Italy)
Abstract: A family of linear regression models is studied where the spatial dependence is modelled by a power series matrix function. In particular, the focus is on a specific choice of functions that are related to probability distribution functions and shows how this family is large enough to encompass some popular models used in the spatial econometric literature, such as SARAR and MESS models. Some insights are provided into the difference between specifications, with emphasis on advantages and shortcomings as well as on the interpretation of the parameters and correspondences between models. The quasi-maximum likelihood estimator is defined and its asymptotic properties under Gaussian and non-Gaussian errors are studied.
