Abstract:
In this paper we carry out a broad simulation experiment where the finite sample behaviour of two parametric seasonal unit root tests for monthly data is studied. In concrete, we analyse the performance of both the Franses (1991) and Canova and Hansesn (1995) procedures. We point out, among other facts, the large distortions in size caused by MAs structures or by misspecification of the seasonal terms. We also show that the tests we study seem to be robust in front periodic seasonality. In this case, the problems are focused on the zero frequency although a decrease in power is also observed in the seasonal frequencies related with the variation of the periodic coefficients. Finally, we point out several facts not expected when we designed the simulation experiments that need future research.
More papers in Working Papers in Economics from Universitat de Barcelona. Espai de Recerca en Economia Address: Espai de Recerca en Economia, Facultat de Ciències Econòmiques. Tinent Coronel Valenzuela, Num 1-11 08034 Barcelona. Spain. Contact information at EDIRC. Series data maintained by Espai de Recerca en Economia ().
This site is part of RePEc
and all the data displayed here is part of the RePEc data set.
Is your work missing from RePEc? Here is how to
contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .