GM estimation of higher order spatial autoregressive processes in panel data error component models

This paper presents a generalized moments (GM) approach to estimating an R-th order spatial regressive process in a panel data error component model. We derive moment conditions to estimate the parameters of the higher order spatial regressive process and the optimal weighting matrix required to achieve asymptotic efficiency. We prove consistency of the proposed GM estimator and provide Monte Carlo evidence that it performs well also in reasonably small samples.