Political Analysis Advance Access originally published online on February 24, 2007
Political Analysis 2007 15(3):347-363; doi:10.1093/pan/mpm004
| ||||||||||||||||||||||||||||||||||||||||||||||||||
A Simple Distribution-Free Test for Nonnested Model Selection
Department of Political Science, University of Rochester, Rochester, NY 14627-0146
e-mail: kevin.clarke{at}rochester.edu
This paper considers a simple distribution-free test for nonnested model selection. The new test is shown to be asymptotically more efficient than the well-known Vuong test when the distribution of individual log-likelihood ratios is highly peaked. Monte Carlo results demonstrate that for many applied research situations, this distribution is indeed highly peaked. The simulation further demonstrates that the proposed test has greater power than the Vuong test under these conditions. The substantive application addresses the effect of domestic political institutions on foreign policy decision making. Do domestic institutions have effects because they hold political leaders accountable, or do they simply promote political norms that shape elite bargaining behavior? The results indicate that the latter model has greater explanatory power.
Authors' note: This work was supported by National Science Foundation Grant SES-0213771. I thank Paul K. Huth and Todd L. Allee for graciously sharing their data and code. I also thank Bear Braumoeller, Curtis Signorino, Tasos Kalandrakis, participants in the NorthEast Methodology Program, New York University, 2003, and the reviewers for their comments. Errors remain my own. Supplementary materials are available on the Political Analysis Web site.