A0680
Title: Robust simultaneous detection of multiple breaks
Authors: Peiyun Jiang - Tokyo Metropolitan University (Japan) [presenting]
Yoshimasa Uematsu - Hitotsubashi University (Japan)
Abstract: A novel multiple testing methodology is proposed for detecting structural changes in each variable of high-dimensional time series data. The approach enables break tests in the trend functions of individual time series without requiring prior knowledge of whether the noise components are stationary or integrated. The new multiple testing procedure controls the false discovery rate (FDR) at a predetermined level and accounts for complex dependencies in both the time-series and cross-sectional dimensions. The proposed method provides a robust framework for detecting structural breaks in high-dimensional time series settings.
