A Test of M Structural Breaks Under the Unit Root Hypothesis

In this paper we provide tests for the unit root hypothesis against the occurrence of an unspecified number of breaks which may be larger than 2 but smaller than the maximum allowed number of breaks m in univariate time series models. The advocated procedure is computationally efficient and less intensive than those previously suggested in the literature. We further provide critical values for the test and apply the new test to the Nelson and Plosser macroeconomic series.