PANEL COINTEGRATION: ASYMPTOTIC AND FINITE SAMPLE PROPERTIES OF POOLED TIME SERIES TESTS WITH AN APPLICATION TO THE PPP HYPOTHESIS

We examine properties of residual-based tests for the null of no cointegration for dynamic panels in which both the short-run dynamics and the long-run slope coefficients are permitted to be heterogeneous across individual members of the panel. The tests also allow for individual heterogeneous fixed effects and trend terms, and we consider both pooled within dimension tests and group mean between dimension tests. We derive limiting distributions for these and show that they are normal and free of nuisance parameters. We also provide Monte Carlo evidence to demonstrate their small sample size and power performance, and we illustrate their use in testing purchasing power parity for the post–Bretton Woods period.I thank Rich Clarida, Bob Cumby, Mahmoud El-Gamal, Heejoon Kang, Chiwha Kao, Andy Levin, Klaus Neusser, Masao Ogaki, David Papell, Pierre Perron, Abdel Senhadji, Jean-Pierre Urbain, Alan Taylor, and three anonymous referees for helpful comments on various earlier versions of this paper. The paper has also benefited from presentations at the 1994 North American Econometric Society Summer Meetings in Quebec City, the 1994 European Econometric Society Summer Meetings in Maastricht, and workshop seminars at the Board of Governors of the Federal Reserve, INSEE-CREST Paris, IUPUI, Ohio State, Purdue, Queens University Belfast, Rice University–University of Houston, and Southern Methodist University. Finally, I thank the following students who provided assistance in the earlier stages of the project: Younghan Kim, Rasmus Ruffer, and Lining Wan.