讲座题目:Change point detection for tensors with heterogeneous slices
主讲人:朱力行
讲座时间:2023年12月7日(周四)14:30-15:30
讲座地点:国际会议中心一楼
主讲人简介:
讲座摘要:
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.