Panel Unit Roots Tests for Cross-Sectionally Correlated Panels: A Monte Carlo Comparison

This paper deals with the finite sample performance of a set of unit root tests for cross correlated panels. As is well known, univariate tests are not powerful to reject the null of a unit root for the usual economic variables while panel tests, by exploiting the large number of cross-section units, provide a device to increase the power of unit root tests. We investigate the finite sample properties of recently proposed panel unit root tests for cross-sectionally correlated panels. Specifically, the size and power of Choi’s (2002), Bai and Ng’s (2003), Moon and Perron’s (2003), and Phillips and Sul’s (2003) tests are analyzed by a Monte Carlo simulation study. In synthesis, Moon and Perron’s (2003) tests show good size and power for different values of T and N and model specifications. Focusing on Bai and Ng’s (2003) procedure, the simulation study highlights first that the suggested ADF test for the nonstationary analysis of the common factor lack of power, and secondly the simulation shows that the pooled Dickey-Fuller-GLS test provides higher power than the pooled ADF test for the analysis of nonstationary properties of the idiosyncratic components. Choi’s (2002) tests are strongly oversized when the common factor influences the cross-section units heterogeneously. Finally, all the tests lack power when a deterministic trend is included in the data generating process.