The main aim of this paper is to compare the size and size-adjusted power prop-erties of four residual-based and one maximum-likelihood-based panel cointegrationtests with the help of Monte Carlo simulations. In this study the panel-rho, the group-rho, the panel-t, the group-t statistics of Pedroni (1999) and the standardized LR-barstatistic of Larsson et al. (2001) are considered. The simulation results indicate thatthe panel-t and standardized LR-bar statistic have the best size and power proper-ties among the five panel cointegration test statistics evaluated. Finally, the FisherHypothesis is tested with two different data sets for OECD countries. The resultspoint out the existence of the Fisher relation.
C15 - Statistical Simulation Methods; Monte Carlo Methods ; C23 - Models with Panel Data ; C33 - Models with Panel Data ; Business administration. Other aspects ; Individual Working Papers, Preprints ; No country specification