Comparison of unit root tests for time series with level shifts

Abstract

Unit root tests are considered for time series which have a level shift at a known point in time. The shift can have a very general nonlinear form, and additional deterministic mean and trend terms are allowed for. Prior to the tests, the deterministic parts and other nuisance parameters of the data generation process are estimated in a first step. Then, the series are adjusted for these terms and unit root tests of the Dickey–Fuller type are applied to the adjusted series. The properties of previously suggested tests of this sort are analysed and modifications are proposed which take into account estimation errors in the nuisance parameters. An important result is that estimation under the null hypothesis is preferable to estimation under local alternatives. This contrasts with results obtained by other authors for time series without level shifts.

Islamic Banking, Credit and Economic Growth: Some Empirical Evidence

“…On the basis of the trace and eigenvalues statistics (see Table 5, Panel A), the null of no cointegration cannot be rejected at the 5% level only in the case Guatemala among the countries without Islamic banks; therefore it appears that there is a stable long-run relationship between credit and GDP almost in every case. As for countries with Islamic banks, both the trace and eigenvalue statistics reject the null hypothesis of no cointegration at the 5% level for all but of one of them, namely Turkey, for which the results are contradictory (see Panel B) -in this case we give more weight to the trace statistic that suggests cointegration, because this test is known to provide more robust results than the maximal eigenvalues one (see Luintel andKhan, 1999, andLanne et al, 2002). (Johansen, 1995).…”

The contribution of domestic, regional and international factors to Latin America's business cycle

“…Furthermore, in order to take into account the possibility of structural breaks due to …nancial crises and recessions, we performed the ADF unit root test with breaks proposed by Saikkonen and Lütkepohl (2002) and Lanne et al (2002Lanne et al ( , 2003. The results are reported in Table 3 [ Table 3 about here]…”

Interdependência dos preços do milho no sul brasileiro

“…O emprego desse teste requer que seja definido o período da quebra, considerando as fases que antecedem e sucedem à mudança estrutural (Cunha, 2008). Os valores críticos foram obtidos por Lanne et al (2002).…”