Mathematics Colloquia and Seminars

Return to Colloquia & Seminar listing

Change-point analysis is thriving in this big data era to address problems arising in many fields where massive data sequences are collected to study complicated phenomena over time.  It plays an important role in processing these data by segmenting a long sequence into homogeneous parts for follow-up studies.  The observation could be of high dimension or not lying in the Euclidean space, such as network data, which are hard to be characterized by parametric models.  We utilize the inter-point information of the observations and propose a series of non-parametric methods to handle the problem.  In particular, we take into account a pattern caused by the curse of dimensionality so that the proposed methods can cover a wide range of alternatives.  Moreover, we work out ways to analytically approximate the p-values of the test statistics to enable easy type I error control.