Publication Type

Journal Article

Version

acceptedVersion

Publication Date

9-2010

Abstract

This paper presents previous termanext term modified previous termLM test of spatial error components,next term which is shown to be previous termrobustnext term against distributional misspecifications and previous termspatialnext term layouts. The proposed previous termtestnext term differs from the previous termLM testnext term of Anselin (2001) by previous termanext term term in the denominators of the previous termtestnext term statistics. This term disappears when either the previous termerrorsnext term are normal, or the variance of the diagonal elements of the product of previous termspatialnext term weights matrix and its transpose is zero or approaches to zero as sample size goes large. When neither is true, as is often the case in practice, the effect of this term can be significant even when sample size is large. As previous termanext term result, there can be severe size distortions of the Anselin's previous termLM test, anext term phenomenon revealed by the Monte Carlo results of Anselin and Moreno (2003) and further confirmed by the Monte Carlo results presented in this paper. Our Monte Carlo results also show that the proposed previous termtestnext term performs well in general.

Keywords

Distributional misspecifications, LM tests, Robustness, Spatial error components, Spatial layouts

Discipline

Econometrics

Research Areas

Econometrics

Publication

Regional Science and Urban Economics

Volume

40

Issue

5

First Page

299

Last Page

310

ISSN

0166-0462

Identifier

10.1016/j.regsciurbeco.2009.10.001

Publisher

Elsevier

Additional URL

https://doi.org/10.1016/j.regsciurbeco.2009.10.001

Download

Included in

Econometrics Commons

Share

COinS