Frequently, seasonal and non-seasonal data (especially macro time series) are observed with noise. For instance, the time series can have irregular abrupt changes and interruptions following as a result of additive or temporary change outliers caused by external circumstances which are...

This paper studies the asymptotic behaviour of the Dickey-Fuller family of tests when the variable being considered exhibits a break in the seasonal pattern. We show that the Dickey-Fuller test tends to reject the unit root null hypothesis, except when the break affects all the periods in a...

This Phillips-Ouliaris (1988) non-parametric unit root test for non seasonal data and the seasonal one of Joyeux (1992) are based in the estimation of the spectral density function in a fixed frequency. We can get consistent estimations of the spectrum using spectral windows, but such procedure...

The lack of suitable critical values for the Dickey-Fuller integrability test in finite-samples can drive researchers to spurious conclusions when using asymptotic critical values. In this paper we estimate response surfaces for the Dickey-Fuller unit root test with structural breaks that allow...

This paper analyses the role of shocks in Spanish economic growth over the period 1850-1990. In the existence of a unit root, the trend is stochastic, which implies that the series has a long memory, and shocks have persistent effects. As a result, the series does not return to its former path...

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...