Generalized shifted-factor analysis method for multivariate geo-referenced data

Multivariate data with spatial dependencies arise in many areas of application, including geology, precision agriculture, and ecology. For analysis of such data, a methodology based on a generalized shifted-factor model is developed. The model incorporates potential lagged dependencies between factors and observed variables, representing asymmetric spatial dependencies observed in practice. Identification and estimation issues are discussed. A prediction procedure that exploits both the multivariate and spatial dependence in the data is proposed and illustrated.