Experimental and Empirical Investigations on the Spatial Error Model

点击翻译按钮获取中文摘要

A regression model is called a spatial error model when spatial autocorrelation is present in the disturbance. A spatial error model is classified into one of two types on the basis of the method of specification of spatial autocorrelation. The first type is defined using a correlation function, which is a device to specify the autocorrelation in a direct manner, whereas the second type is defined using a weight matrix, which is a device to specify it in an indirect manner. The purpose of the present study is to evaluate the performance of spatial error models using experimental and empirical approaches. As errors cannot be assumed to be absent in any specification, choosing a relatively well-performing model is a good policy in order to allow for such errors. Methodological issues are addressed with regard to both estimation and prediction, and some results are presented that support the combined use of the two types of models.