04/18/2023
By Hsien-Yuan Hsu
Please join the Social and Health Statistics Seminar on April 19, 2023 at noon in Coburn 275. This series of seminars is co-sponsored by the Zuckerberg College of Health Sciences, the Center for Health Statistics, and the College of Fine Arts, Humanities, & Social Sciences.
Title: Bayesian Change-Point Detection in High Dimension: Recent Developments and Applications
Speaker: Nilabja Guha, assistant professor in the Department of Mathematics and Statistics at UMass Lowell.
Abstract
Many dynamic and random processes in nature go through sudden and significant structural changes. Often the change is in the observable quantity in response to a change in a latent factor. Such ‘change-points’ are routinely observed across all scientific disciplines and applications, such as economics, epidemiology, social sciences, cybersecurity and finance. Specific examples could be changing regression when the observed variable depends on predictors through a mean structure that changes with time, or change points in data with massive dimensions, such as high-resolution imaging data or complex connected graphs. While there is a substantial literature proposing elaborate methods for detecting change points in different settings, there has been limited consideration of Bayesian methods for change-points that can account for hierarchical models with complex dependence or sparsity structures. This work fills this gap with new statistical tools motivated by real-life applications, by developing a new theoretical framework while retaining efficiency and usefulness in current applications. Here we present our current developments and explore the scope of interdisciplinary applications.
Short Bio
Guha's research interests include general high-dimensional problems, Bayesian modeling, change point estimation, density and function estimation, inverse problem and uncertainty quantification etc. As the PI of an NSF grant (DMS) `New Directions in Bayesian Change-Point Analysis', he is currently developing scalable change-point detection methodologies in high dimensional setup with applications in health sciences, economics, finance etc. He is also a Co-PI of a Spencer Foundation grant on education that studies the effect of student mobility on education, using Bayesian Growth Mixture Models (GMM).
Before joining UMass Lowell he was a postdoctoral research associate in the Department of Statistics and the Institute for Scientific Computation at Texas A&M University. He received his Ph.D. in Statistics from the University of Maryland Baltimore County.