A0371
Title: Robust simultaneous detection of multiple breaks
Authors: Yoshimasa Uematsu - Hitotsubashi University (Japan) [presenting]
Peiyun Jiang - Tokyo Metropolitan University (Japan)
Abstract: A novel methodology is proposed for detecting structural changes in each variable of high-dimensional time series. The structural breaks are detected by multiple testing while controlling the false discovery rate (FDR) at a predetermined level. However, there are two main challenges. The first is that, due to the high dimensionality of the time series, it is difficult to determine in advance whether each variable is I(0) or I(1). The second challenge arises from the complex dependencies in both the time series and cross-sectional dimensions, which prevent the use of conventional p-values from guaranteeing FDR control. The multiple testing methodology addresses these issues.
