Time: 4:30-5:30 p.m.
Date: Wednesday, September 24, 2025
Speaker: Kangjie Zhou, Founder's Postdoctoral Fellow, Department of Statistics, Columbia University
Title: Dynamic Factor Analysis of High-dimensional Recurrent Events
Abstract: Recurrent event time data arise in many studies, including biomedicine, public health, marketing, and social media analysis. High-dimensional recurrent event data involving large numbers of event types and observations become prevalent with the advances in information technology. In this talk, we propose a semiparametric dynamic factor model for the dimension reduction and prediction of high-dimensional recurrent event data. The proposed model imposes a low-dimensional structure on the mean intensity functions of the event types while allowing for dependencies. For this model, we propose a nearly rate-optimal smoothing-based estimator, and develop an information criterion that consistently selects the number of factors. Simulation studies demonstrate the effectiveness of these inference tools. We further apply the proposed method to grocery shopping data, for which we also obtain an interpretable factor structure. Based on joint work with Fangyi Chen, Yunxiao Chen and Zhiliang Ying.
Bio: Kangjie Zhou is a Founder's Postdoctoral Fellow at Department of Statistics, Columbia University. In 2024, he received PhD in Statistics from Stanford University, where he was advised by Andrea Montanari. His research interests include high-dimensional statistics, non-convex optimization, probability theory, and deep learning.