Ramalho, Joaquim J. S.: Generalized Empirical Likelihood Non-Nested Tests
World Conference Econometric Society, 2000, Seattle

Joaquim J. S. Ramalho, University of Bristol
Richard J. Smith, University of Bristol
Generalized Empirical Likelihood Non-Nested Tests
Session: C-11-15  Tuesday 15 August 2000  by Ramalho, Joaquim J. S.
This paper examines non-nested tests for competing moment conditions using a semi-parametric generalized empirical likelihood framework. By associating an additional parameter for each of the moment conditions, the generalized empirical likelihood method embeds sample versions of the moment conditions used in generalized method of moments (GMM) estimation in a non-parametric likelihood-type function. The resulting estimators from optimising the generalized empirical likelihood criteria are first order asymptotically equivalent to those based on GMM. Cox-type, moment encompassing and parametric encompassing non-nested tests for competing moment condition models are proposed. A simulation experiment is conducted to examine the efficacy of the proposed statistics in terms of their size and power properties and to compare their properties with those of corresponding non-nested test statistics based on GMM estimation.


File created by Jurgen Doornik with eswc2000.ox on 2-01-2001