Applied Mathematics Colloquia by Xiaofeng Shao - Change Point Detection via Self-normalization: A Personal Journey

Speaker: Xiaofeng Shao, University of Illinois

Title: Change Point Detection via Self-normalization: A Personal Journey 

Abstract:  Change point detection is a classical topic in Statistics and has many applications. Motivated by the occurrence of structural breaks in data of growing dimension from areas such as genomics, finance and neuroscience, there is a surge of interest in change point detection for high-dimensional data with complex cross-sectional and temporal dependence.  

In this talk, we will present a new nonparametric approach to testing and estimation of change points in time series data via self-normalization.  The talk will start with a review of self-normalization for inference of time series in the context of confidence interval construction and change-point testing in mean. Then we will briefly discuss some recently published work on segmenting Covid-19 time series and high-dimensional change-point detection. Finally, we will focus on dimension-agnostic change point testing, where the proposed test can work for both low and high-dimensional settings and accommodate both temporal and cross-sectional dependence.  

Bio: Prof. Xiaofeng Shao received his PhD degree in Statistics from the University of Chicago in 2006 and has since been a faculty member with the Department of Statistics at the University of Illinois Urbana-Champaign. His current research interests include time series analysis, change-point analysis, functional data analysis, high dimensional data analysis and their applications. He is a fellow of Institute of Mathematical Statistics (IMS) and American Statistical Association (ASA). He currently serves as an associate editor for Journal of Royal Statistical Society, Series B, Journal of the American Statistical Association and Journal of Time Series Analysis.

Applied Mathematics Colloquia