Title: Multiple change-point detection under short-range dependence
Authors: Sara Kristin Schmidt - Ruhr-University Bochum (Germany) [presenting]
Herold Dehling - Ruhr-University Bochum (Germany)
Roland Fried - TU Dortmund University (Germany)
Max Wornowizki - TU Dortmund University (Germany)
Abstract: A new test statistic is introduced to examine the presence of, potentially multiple, change-points in the variance of an absolutely regular time series. By regarding the test statistic as a U-statistic of a triangular array, a law of large numbers and a central limit theorem are established. Afterwards, asymptotic critical values of the test statistic are obtained and its performance is evaluated and compared to alternatives from the literature via a simulation study.