Nonparametric likelihood ratio model selection tests between parametric likelihood and moment condition models
- Creators
- Che, Xiaohong
- Hong, Han
-
Shum, Matthew
Abstract
We propose a nonparametric likelihood ratio testing procedure for choosing between a parametric (likelihood) model and a moment condition model when both models could be misspecified. Our procedure is based on comparing the Kullback–Leibler Information Criterion (KLIC) between the parametric model and moment condition model. We construct the KLIC for the parametric model using the difference between the parametric log likelihood and a sieve nonparametric estimate of population entropy, and obtain the KLIC for the moment model using the empirical likelihood statistic. We also consider multiple (>2) model comparison tests, when all the competing models could be misspecified, and some models are parametric while others are moment-based. We evaluate the performance of our tests in a Monte Carlo study, and apply the tests to an example from industrial organization.
Additional Information
© 2007 Elsevier. Available online 8 March 2007. We thank the associate editor, three referees, J. Geweke, J. Horowitz, Y. Kitamura, O. Linton, N. Meddahi, S. Ng, H. Paarsch, R. Porter and seminar participants in Carnegie-Mellon, U. Iowa, the 2003 Econometric Society Winter Meetings in Washington DC, SITE 2003, and the 2004 Semiparametrics conference in Rio for helpful comments. We also thank D. Chen and in particular D. Nekipelov for excellent research assistance in the simulation study. All three authors gratefully acknowledge supports from the National Science Foundation, and Hong also acknowledges support from the Sloan Foundation.Additional details
- Eprint ID
- 12059
- Resolver ID
- CaltechAUTHORS:CHEje07
- NSF
- Alfred P. Sloan Foundation
- Created
-
2008-10-22Created from EPrint's datestamp field
- Updated
-
2023-06-01Created from EPrint's last_modified field