Bootstrap inference in spatial econometrics : the J test

B. Fingleton, P. Burridge

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

Kelejian (2008) introduces a J-type test for the situation in which a null linear regression model, Model0, is to be tested against one or more rival non-nested alternatives, Model1, . . ., Modelg, where typically the competing models possess endogenous spatial lags and spatially autoregressive error processes. Concentrating on the case g1, in this paper we examine the finite sample properties of a spatial J statistic that is asymptotically x2 2 under the null, and an alternative version that is conjectured to be approximately x2 1; both introduced by Kelejian. We demonstrate numerically that the tests are excessively liberal in some leading cases and conservative in others using the relevant chi-square asymptotic approximations, and explore how far this may be corrected using a simple bootstrap resampling method.
LanguageEnglish
Pages93-119
Number of pages26
JournalSpatial Economic Analysis
Volume5
Issue number1
DOIs
Publication statusPublished - Mar 2010

Fingerprint

Econometrics
econometrics
Bootstrap
Null
Resampling Methods
bootstrapping
Chi-square
Alternatives
Bootstrap Method
Statistic
statistics
Demonstrate
test
Spatial econometrics
Bootstrap inference
Model
Finite sample properties
Spatial lag
Statistics
Resampling methods

Keywords

  • spatial econometrics
  • bootstrap
  • J-test

Cite this

Fingleton, B. ; Burridge, P. / Bootstrap inference in spatial econometrics : the J test. In: Spatial Economic Analysis. 2010 ; Vol. 5, No. 1. pp. 93-119.
@article{c01137b13c6b45829ec09a2a1855034c,
title = "Bootstrap inference in spatial econometrics : the J test",
abstract = "Kelejian (2008) introduces a J-type test for the situation in which a null linear regression model, Model0, is to be tested against one or more rival non-nested alternatives, Model1, . . ., Modelg, where typically the competing models possess endogenous spatial lags and spatially autoregressive error processes. Concentrating on the case g1, in this paper we examine the finite sample properties of a spatial J statistic that is asymptotically x2 2 under the null, and an alternative version that is conjectured to be approximately x2 1; both introduced by Kelejian. We demonstrate numerically that the tests are excessively liberal in some leading cases and conservative in others using the relevant chi-square asymptotic approximations, and explore how far this may be corrected using a simple bootstrap resampling method.",
keywords = "spatial econometrics, bootstrap, J-test",
author = "B. Fingleton and P. Burridge",
year = "2010",
month = "3",
doi = "10.1080/17421770903511346",
language = "English",
volume = "5",
pages = "93--119",
journal = "Spatial Economic Analysis",
issn = "1742-1772",
number = "1",

}

Bootstrap inference in spatial econometrics : the J test. / Fingleton, B.; Burridge, P.

In: Spatial Economic Analysis, Vol. 5, No. 1, 03.2010, p. 93-119.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Bootstrap inference in spatial econometrics : the J test

AU - Fingleton, B.

AU - Burridge, P.

PY - 2010/3

Y1 - 2010/3

N2 - Kelejian (2008) introduces a J-type test for the situation in which a null linear regression model, Model0, is to be tested against one or more rival non-nested alternatives, Model1, . . ., Modelg, where typically the competing models possess endogenous spatial lags and spatially autoregressive error processes. Concentrating on the case g1, in this paper we examine the finite sample properties of a spatial J statistic that is asymptotically x2 2 under the null, and an alternative version that is conjectured to be approximately x2 1; both introduced by Kelejian. We demonstrate numerically that the tests are excessively liberal in some leading cases and conservative in others using the relevant chi-square asymptotic approximations, and explore how far this may be corrected using a simple bootstrap resampling method.

AB - Kelejian (2008) introduces a J-type test for the situation in which a null linear regression model, Model0, is to be tested against one or more rival non-nested alternatives, Model1, . . ., Modelg, where typically the competing models possess endogenous spatial lags and spatially autoregressive error processes. Concentrating on the case g1, in this paper we examine the finite sample properties of a spatial J statistic that is asymptotically x2 2 under the null, and an alternative version that is conjectured to be approximately x2 1; both introduced by Kelejian. We demonstrate numerically that the tests are excessively liberal in some leading cases and conservative in others using the relevant chi-square asymptotic approximations, and explore how far this may be corrected using a simple bootstrap resampling method.

KW - spatial econometrics

KW - bootstrap

KW - J-test

UR - http://www.scopus.com/inward/record.url?scp=77951533857&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1080/17421770903511346

U2 - 10.1080/17421770903511346

DO - 10.1080/17421770903511346

M3 - Article

VL - 5

SP - 93

EP - 119

JO - Spatial Economic Analysis

T2 - Spatial Economic Analysis

JF - Spatial Economic Analysis

SN - 1742-1772

IS - 1

ER -