Properties of tests for spatial dependence in linear regression models

L. Anselin, S. Rey

Research output: Contribution to journalArticle

357 Citations (Scopus)

Abstract

Based on a large number of Monte Carlo simulation experiments on a regular lattice, we compare the properties of Moran's I and Lagrange multiplier tests for spatial dependence, that is, for both spatial error autocorrelation and for a spatially lagged dependent variable. We consider both bias and power of the tests for six sample sizes, ranging from 25 to 225 observations, for different structures of the spatial weights matrix, for several underlying error distributions, for misspecified weights matrices, and for the situation where boundary effects are present. The results provide an indication of the sample sizes for which the asymptotic properties of the tests can be considered to hold. They also illustrate the power of the Lagrange multiplier tests to distinguish between substantive spatial dependence (spatial lag) and spatial dependence as a nuisance (error autocorrelation). -Authors

Original languageEnglish (US)
Pages (from-to)112-131
Number of pages20
JournalGeographical Analysis
Volume23
Issue number2
StatePublished - 1991
Externally publishedYes

Fingerprint

regression
multiplier
autocorrelation
matrix
indication
simulation
test
experiment
trend
distribution
nuisance
effect

ASJC Scopus subject areas

  • Geography, Planning and Development

Cite this

Properties of tests for spatial dependence in linear regression models. / Anselin, L.; Rey, S.

In: Geographical Analysis, Vol. 23, No. 2, 1991, p. 112-131.

Research output: Contribution to journalArticle

Anselin, L. ; Rey, S. / Properties of tests for spatial dependence in linear regression models. In: Geographical Analysis. 1991 ; Vol. 23, No. 2. pp. 112-131.
@article{9de1a04574764f89aaa6c1a102775803,
title = "Properties of tests for spatial dependence in linear regression models",
abstract = "Based on a large number of Monte Carlo simulation experiments on a regular lattice, we compare the properties of Moran's I and Lagrange multiplier tests for spatial dependence, that is, for both spatial error autocorrelation and for a spatially lagged dependent variable. We consider both bias and power of the tests for six sample sizes, ranging from 25 to 225 observations, for different structures of the spatial weights matrix, for several underlying error distributions, for misspecified weights matrices, and for the situation where boundary effects are present. The results provide an indication of the sample sizes for which the asymptotic properties of the tests can be considered to hold. They also illustrate the power of the Lagrange multiplier tests to distinguish between substantive spatial dependence (spatial lag) and spatial dependence as a nuisance (error autocorrelation). -Authors",
author = "L. Anselin and S. Rey",
year = "1991",
language = "English (US)",
volume = "23",
pages = "112--131",
journal = "Geographical Analysis",
issn = "0016-7363",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Properties of tests for spatial dependence in linear regression models

AU - Anselin, L.

AU - Rey, S.

PY - 1991

Y1 - 1991

N2 - Based on a large number of Monte Carlo simulation experiments on a regular lattice, we compare the properties of Moran's I and Lagrange multiplier tests for spatial dependence, that is, for both spatial error autocorrelation and for a spatially lagged dependent variable. We consider both bias and power of the tests for six sample sizes, ranging from 25 to 225 observations, for different structures of the spatial weights matrix, for several underlying error distributions, for misspecified weights matrices, and for the situation where boundary effects are present. The results provide an indication of the sample sizes for which the asymptotic properties of the tests can be considered to hold. They also illustrate the power of the Lagrange multiplier tests to distinguish between substantive spatial dependence (spatial lag) and spatial dependence as a nuisance (error autocorrelation). -Authors

AB - Based on a large number of Monte Carlo simulation experiments on a regular lattice, we compare the properties of Moran's I and Lagrange multiplier tests for spatial dependence, that is, for both spatial error autocorrelation and for a spatially lagged dependent variable. We consider both bias and power of the tests for six sample sizes, ranging from 25 to 225 observations, for different structures of the spatial weights matrix, for several underlying error distributions, for misspecified weights matrices, and for the situation where boundary effects are present. The results provide an indication of the sample sizes for which the asymptotic properties of the tests can be considered to hold. They also illustrate the power of the Lagrange multiplier tests to distinguish between substantive spatial dependence (spatial lag) and spatial dependence as a nuisance (error autocorrelation). -Authors

UR - http://www.scopus.com/inward/record.url?scp=0025937011&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025937011&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0025937011

VL - 23

SP - 112

EP - 131

JO - Geographical Analysis

JF - Geographical Analysis

SN - 0016-7363

IS - 2

ER -