Abstract
The panel variant of the KPSS tests developed by Hadri [Hadri, K., 2000, Testing for stationarity in heterogeneous panels. Econometrics Journal, 3, 148–161] for the null of stationarity suffers from size distortions in the presence of cross-section dependence. However, applying the bootstrap methodology, we find that these tests are approximately correctly sized.
Acknowledgements
The authors would like to thank participants at the 13th International Panel Data Conference held in Cambridge, United Kingdom, and an anonymous referee for useful comments and suggestions.