A0282
Title: Change-point inference for high-dimensional heteroscedastic data
Authors: Xiaofeng Shao - University of Illinois at Urbana-Champaign (United States) [presenting]
Abstract: A bootstrap-based test is proposed to detect a mean shift in a sequence of high-dimensional observations with unknown time-varying heteroscedasticity. The proposed test builds on the U-statistic-based approach, targets a dense alternative, and adopts a wild bootstrap procedure to generate critical values. The bootstrap-based test is free of tuning parameters and is capable of accommodating unconditional time-varying heteroscedasticity in high-dimensional observations, as demonstrated in the theory and simulations. Theoretically, the bootstrap consistency is justified by using the recently proposed unconditional approach. Extensions to testing for multiple change points and estimation using wild binary segmentation are also presented. Numerical simulations demonstrate the robustness of the proposed testing and estimation procedures with respect to different kinds of time-varying heteroscedasticity.
