Empirical evidence on the robustness of the weighted symmetric unit root test

Abstract

Recent research has shown the Dickey–Fuller (Econometrica, 49, pp. 1057–72, 1979) test to suffer severe size distortion in the presence of structural breaks under the unit root null hypothesis. In contrast, further theoretical results have suggested the weighted symmetric Dickey–Fuller test to be robust to such breaks. In this study, the empirical results of the seminal study of Leybourne et al. (Journal of Econometrics, 87, pp. 191–203, 1998) are revisited to provide an empirical illustration of the robustness of the weighted symmetric Dickey–Fuller test to structural breaks.