Testing for Non-Nested Conditional Moment Restrictions Using Unconditional Empirical Likelihood

30 Pages Posted: 23 May 2008  

Taisuke Otsu

Yale University - Cowles Foundation

Myung Hwan Seo

Seoul National University - School of Economics

Yoon-Jae Whang

Seoul National University - School of Economics

Date Written: May 16, 2008

Abstract

We propose non-nested hypotheses tests for conditional moment restriction models based on the method of generalized empirical likelihood (GEL). By utilizing the implied GEL probabilities from a sequence of unconditional moment restrictions that contains equivalent information of the conditional moment restrictions, we construct Kolmogorov-Smirnov and Cramer-von Mises type moment encompassing tests. Advantages of our tests over Otsu and Whang's (2007) tests are: (i) they are free from smoothing parameters, (ii) they can be applied to weakly dependent data, and (iii) they allow non-smooth moment functions. We derive the null distributions, validity of a bootstrap procedure, and local and global power properties of our tests. The simulation results show that our tests have reasonable size and power performance in finite samples.

Keywords: Empirical likelihood, Non-nested tests, Conditional moment restrictions

JEL Classification: C12, C13, C14, C22

Suggested Citation

Otsu, Taisuke and Seo, Myung Hwan and Whang, Yoon-Jae, Testing for Non-Nested Conditional Moment Restrictions Using Unconditional Empirical Likelihood (May 16, 2008). Cowles Foundation Discussion Paper No. 1660. Available at SSRN: https://ssrn.com/abstract=1136109

Taisuke Otsu (Contact Author)

Yale University - Cowles Foundation ( email )

Box 208281
New Haven, CT 06520-8281
United States

Myung Hwan Seo

Seoul National University - School of Economics ( email )

San 56-1, Silim-dong, Kwanak-ku
Seoul 151-742

Yoon-Jae Whang

Seoul National University - School of Economics ( email )

San 56-1, Silim-dong, Kwanak-ku
Seoul 151-742
Korea
+82 2 80 6362 (Phone)
+82 2 86 4231 (Fax)

HOME PAGE: http://plaza.snu.ac.kr/~whang

Paper statistics

Downloads
38
Abstract Views
341