题目： Change point detection for tensors with  heterogeneous slices

报告人：朱力行

报告时间：2023年12月7日（周四）14:30-15:30

地点：国际会议中心一楼

报告人简介：

报告摘要：

In many applications, tensor data  may consist of    heterogeneous slices according to a categorical mode, and independent but not identically distributed error tensors over time.    To detect change structures in such tensor data,
we define a mode-based signal-screening  Frobenius distance for the moving sums of slices to handle both dense and sparse model structures of the tensors. Based on this distance, we  construct  a mode-based signal statistic using  a sequence of  ratios with mode-based adaptive-to-change ridge functions.  The  number of changes and their  locations can  be consistently estimated in certain senses, and  the confidence intervals of the locations of change points are constructed when the standardized error tensors are homogenous. The results hold when the size of the tensor and the number of change points diverge at certain rates, respectively. Numerical studies are conducted to examine the finite sample performances of the proposed method. We also analyze two real data examples for illustration.