The Likelihood Ratio Test of Common Factors under Non-Ideal Conditions

The Spatial Durbin model occupies an interesting position in SpatialEconometrics. It is the reduced form of a model with cross-sectional dependencein the errors and it may be used as the nesting equation in a more general approachof model selection. Specifically, in this equation we can obtain the Likelihood Ratiotest of Common Factors (LRCOM). This test has good properties if the model iscorrectly specified, as shown in Mur and Angulo (2006). However, as far as weknow, there is no literature in relation to the behaviour of the test under non-idealconditions, which is the purpose of the paper. Specifically, we study the performanceof the test in the case of heteroscedasticity, non-normality, endogeneity,dense weighting matrices and non-linearity. Our results offer a positive view of theLikelihood Ratio test of Common Factors, which appears to be a useful techniquein the toolbox of spatial econometrics.