Small Sample Properties of Maximum Likelihood Versus Generalized Method of Moments Based Tests for Spatially Autocorrelated Errors

32 Pages Posted: 7 Nov 2005

See all articles by Peter H. Egger

Peter H. Egger

Ifo Institute for Economic Research - International Trade and Foreign Direct Investment; ETH Zürich; Ludwig-Maximilians University of Munich; CESifo (Center for Economic Studies and Ifo Institute for Economic Research)

Mario Larch

University of Bayreuth - Faculty of Law, Business and Economics; University of Bayreuth; CESifo; Ifo Institute for Economic Research; GEP at the University of Nottingham; Centre d'Etudes Prospectives et d'Info. Internationales (CEPII)

Michael Pfaffermayr

University of Innsbruck - Department of Economics; Austrian Institute of Economic Research

Janette Walde

University of Innsbruck

Date Written: October 2005

Abstract

This paper undertakes a Monte Carlo study to compare MLE-based and GMM-based tests regarding the spatial autocorrelation coefficient of the error term in a Cliff and Ord type model. The main finding is that a Wald-test based on GMM estimation as derived by Kelejian and Prucha (2005a) performs surprisingly well. Our Monte Carlo study indicates that the GMM Wald-test is correctly sized even in small samples and exhibits the same power as their MLE-based counterparts. Since GMM estimates are much easier to calculate, the GMM Wald-test is recommended for applied researches.

Keywords: spatial autocorrelation, hypothesis tests, Monte Carlo studies, maximum likelihood estimation, generalized method of moments

JEL Classification: C12, C21, R10

Suggested Citation

Egger, Peter H. and Larch, Mario and Pfaffermayr, Michael and Walde, Janette, Small Sample Properties of Maximum Likelihood Versus Generalized Method of Moments Based Tests for Spatially Autocorrelated Errors (October 2005). CESifo Working Paper No. 1558, Available at SSRN: https://ssrn.com/abstract=843367 or http://dx.doi.org/10.2139/ssrn.843367

Peter H. Egger (Contact Author)

Ifo Institute for Economic Research - International Trade and Foreign Direct Investment ( email )

Poschingerstr. 5
Munich, DE-81679
Germany
+49 0 89 9224 1238 (Phone)
+49 0 89 985369 (Fax)

HOME PAGE: http://www.cesifo-group.de/portal/page?_pageid=36,425628&_dad=portal&_schema=PORTAL

ETH Zürich ( email )

LEE G104
Leonhardstrasse 21
Zurich
Switzerland

Ludwig-Maximilians University of Munich

Schackstr. 4
Munich, 80539
Germany

CESifo (Center for Economic Studies and Ifo Institute for Economic Research)

Poschinger Str. 5
Munich, DE-81679
Germany

Mario Larch

University of Bayreuth - Faculty of Law, Business and Economics ( email )

Universitätsstraße 30
Bayreuth, 95447
Germany
0049 / (0) 921 - 55 6240 (Phone)

University of Bayreuth ( email )

Universitatsstr 30
Bayreuth, D-95447
Germany

CESifo ( email )

Poschinger Str. 5
Munich, DE-81679
Germany

Ifo Institute for Economic Research ( email )

Poschinger Str. 5
Munich, 01069
Germany

GEP at the University of Nottingham ( email )

University Park
Nottingham, NG7 2RD
United Kingdom

Centre d'Etudes Prospectives et d'Info. Internationales (CEPII) ( email )

9 rue Georges Pitard
Paris Cedex 15, F-75015
France

Michael Pfaffermayr

University of Innsbruck - Department of Economics ( email )

Universitaetsstrasse 15
A-6020, Innsbruck
Austria
+43 512 507 7359 (Phone)

HOME PAGE: http://homepage.uibk.ac.at/~c43236/

Austrian Institute of Economic Research ( email )

P.O. Box 91
Wien, A-1103
Austria
+43 1 798 26 01-253 (Phone)
+43 1 798 93 86 (Fax)

HOME PAGE: http://www.wifo.ac.at/Michael.Pfaffermayr/

Janette Walde

University of Innsbruck ( email )

Universitatsstrasse 15
Innsbruck, A-6020
Austria