B0451
Title: On distributional change-point detection in functional time series
Authors: Lajos Horvath - University of Utah (USA)
B Cooper Boniece - Drexel University (United States) [presenting]
Lorenzo Trapani - University of Leicester (United Kingdom)
Abstract: The problem of testing for distributional changes in functional time series data is considered. The approach is based on statistics related to the empirical energy distance. Under a flexible weak temporal dependence assumption that encompasses several time series models, the asymptotic size and consistency of the method are established. Some applications are presented. Time permitting, some related methods for high-dimensional time series are discussed.
