Kolloquium: Annika Betken (University of Twente Enschede): „Testing for Independence Between Long-Range Dependent Time Series Based on Distance Correlation“

Testing for Independence Between Long-Range Dependent Time Series Based on Distance Correlation – Vortragende: Annika Betken, University of Twente Enschede – Einladende: M.-C. Düker

Abstract: To test for independence of two time series we propose a linear combination of empirical distance cross-covariances as test statistic. Its asymptotic distribution is derived under the hypothesis of independence, while consistency is shown to hold for a large class of alternatives. We observe that, depending on the dependence in the data, standardization and limit distributions of the proposed test statistic vary. Since the limit distributions are unknown, we establish a subsampling procedure to determine the critical values for the proposed test.