GMM estimation of the autoregressiveparameter in a spatial autoregressive errormodel using regression residuals

This paper suggests an improved GMM estimator for the autoregressive parameter of a spatial autoregressive error model by taking into account that unobservable regression disturbances are di.erent from observable regression residuals. Although this di.erence decreases in large samples, it is important in small samples. Monte Carlo simu­lations show that the bias can be reduced by 65 - 80% compared to a GMM estimator that neglects the difference between disturbances and residuals. The mean squared error is smaller, too.