Testing the exogeneity assumption in panel data models with "non classical" disturbances
Testing the exogeneity assumption in panel data models with "non classical" disturbances
This paper is concerned with the use of the Durbin-Wu-Hausman test for correlated effects with panel data. The assumptions underlying the construction of the statistic are too strong in many empirical cases. The consequences of deviations from the basic assumptions are investigated. The size distortion is assessed. In the case of measurement error, the Hausman test is found to be a test of the difference in asymptotic biases of between and within group estimators. However, its 'size' is sensitive to the relative magnitude of the intra-group and inter-group variations of the covariates, and can be so large as to preclude the use of the statistic in this case. We show to what extent some assumptions can be relaxed in a panel data context and we discuss an alternative robust formulation of the test. Power considerations are presented.
models with panel data, hausman test, minimum variance estimators, quadratic forms in normal variables, monte carlo simulations
University of Southampton
O'Brien, Raymond
6d46f2be-6f1d-4bcd-9b94-baedee23ff22
Patacchini, Eleonora
42a2cbc9-016c-43f2-a9e9-e2f00172d919
1 February 2003
O'Brien, Raymond
6d46f2be-6f1d-4bcd-9b94-baedee23ff22
Patacchini, Eleonora
42a2cbc9-016c-43f2-a9e9-e2f00172d919
O'Brien, Raymond and Patacchini, Eleonora
(2003)
Testing the exogeneity assumption in panel data models with "non classical" disturbances
(Discussion Papers in Economics and Econometrics, 302)
Southampton, UK.
University of Southampton
68pp.
Record type:
Monograph
(Discussion Paper)
Abstract
This paper is concerned with the use of the Durbin-Wu-Hausman test for correlated effects with panel data. The assumptions underlying the construction of the statistic are too strong in many empirical cases. The consequences of deviations from the basic assumptions are investigated. The size distortion is assessed. In the case of measurement error, the Hausman test is found to be a test of the difference in asymptotic biases of between and within group estimators. However, its 'size' is sensitive to the relative magnitude of the intra-group and inter-group variations of the covariates, and can be so large as to preclude the use of the statistic in this case. We show to what extent some assumptions can be relaxed in a panel data context and we discuss an alternative robust formulation of the test. Power considerations are presented.
Text
0302.pdf
- Version of Record
More information
Published date: 1 February 2003
Keywords:
models with panel data, hausman test, minimum variance estimators, quadratic forms in normal variables, monte carlo simulations
Identifiers
Local EPrints ID: 33202
URI: http://eprints.soton.ac.uk/id/eprint/33202
ISSN: 0966-4246
PURE UUID: 31095536-9315-43a5-b80f-6b6359e3d078
Catalogue record
Date deposited: 18 May 2006
Last modified: 15 Mar 2024 07:43
Export record
Contributors
Author:
Eleonora Patacchini
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics