CMStatistics 2017: Start Registration
View Submission - CFE
Title: Wavelet variance ratio test and wavestrapping for the determination of the cointegration rank Authors:  Burak Alparslan Eroglu - Istanbul Bilgi University (Turkey) [presenting]
Abstract: A wavelet based cointegration test for fractionally integrated time series is proposed. The test is non-parametric and asymptotically invariant to different forms of short run dynamics. The use of wavelets allows one to take advantage of the wavelet based bootstrapping method particularly known as wavestrapping. In this regard, we introduce a new wavestrapping algorithm for multivariate time series processes, specifically for cointegration tests. The Monte Carlo simulations indicate that this new wavestrapping procedure can alleviate the severe size distortions which are generally observed in cointegration tests with time series containing innovations with highly negative MA parameters. Further, we apply the the proposed methodology to analyze long run co-movements in the Credit Default Swap market of European Union Countries.