Nonparametric testing for exogeneity with discrete regressors and instruments
Nonparametric testing for exogeneity with discrete regressors and instruments
This paper presents new approaches to testing for exogeneity in non-parametric models with discrete regressors and instruments. Our interest is in learning about an unknown structural (conditional mean) function. An interesting feature of these models is that under endogeneity the identifying power of a discrete instrument depends on the number of support points of the instruments relative to that of the regressors, a result driven by the discreteness of the variables. Observing that the simple nonparametric additive error model can be interpreted as a linear regression, we present two test-statistics. For the point identifying model, the test is an adapted version of the standard Wu-Hausman approach. This extends the work of Blundell and Horowitz (2007) to the case of discrete regressors and instruments. For the set identifying model, the Wu-Hausman approach is not available. In this case the test-statistic is derived from a constrained minimization problem. The asymptotic distributions of the test-statistics are derived under the null and fixed and local alternatives. The tests are shown to be consistent, and a simulation study reveals that the proposed tests have satisfactory finite-sample properties.
Centre for Microdata Methods and Practice
Hillier, Grant
3423bd61-c35f-497e-87a3-6a5fca73a2a1
Bech, Katarzyna
3d520a54-dd93-4336-b008-99c53861ab45
March 2015
Hillier, Grant
3423bd61-c35f-497e-87a3-6a5fca73a2a1
Bech, Katarzyna
3d520a54-dd93-4336-b008-99c53861ab45
Hillier, Grant and Bech, Katarzyna
(2015)
Nonparametric testing for exogeneity with discrete regressors and instruments
(Cemmap Working Papers, CWP 11/15)
London, GB.
Centre for Microdata Methods and Practice
(doi:10.1920/wp.cem.2015.1115).
Record type:
Monograph
(Discussion Paper)
Abstract
This paper presents new approaches to testing for exogeneity in non-parametric models with discrete regressors and instruments. Our interest is in learning about an unknown structural (conditional mean) function. An interesting feature of these models is that under endogeneity the identifying power of a discrete instrument depends on the number of support points of the instruments relative to that of the regressors, a result driven by the discreteness of the variables. Observing that the simple nonparametric additive error model can be interpreted as a linear regression, we present two test-statistics. For the point identifying model, the test is an adapted version of the standard Wu-Hausman approach. This extends the work of Blundell and Horowitz (2007) to the case of discrete regressors and instruments. For the set identifying model, the Wu-Hausman approach is not available. In this case the test-statistic is derived from a constrained minimization problem. The asymptotic distributions of the test-statistics are derived under the null and fixed and local alternatives. The tests are shown to be consistent, and a simulation study reveals that the proposed tests have satisfactory finite-sample properties.
Text
cwp111515.pdf
- Other
More information
Published date: March 2015
Organisations:
Economics
Identifiers
Local EPrints ID: 375185
URI: http://eprints.soton.ac.uk/id/eprint/375185
PURE UUID: 9cfd74f8-cad6-4010-bdde-3b7a83b037b0
Catalogue record
Date deposited: 18 Mar 2015 15:17
Last modified: 15 Mar 2024 02:44
Export record
Altmetrics
Contributors
Author:
Katarzyna Bech
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