时 间：2022年12月29日（周四）16：00-17：00
地 点：腾讯会议：588 869 9283
主 题：Compound Decision for Parallel Sequential Change Detection
主讲人：Yunxiao Chen London School of Economics and Political Science助理教授
主持人：张思亮 助理教授
主 办：统计学院
摘 要：
This talk will introduce the problem of parallel sequential change detection, which receives wide real-world applications in education, marketing, personalised medicine, and cloud computing, among many others. This problem concerns detecting change points in parallel data streams, where each stream has its own change point, at which its data has a distributional change. With sequentially observed data, a decision-maker needs to declare whether changes have already occurred to the streams at each time point. Once a stream is declared to have changed, the decision-maker will intervene, such as deactivating the stream so that its future data will no longer be collected. We argue that for many applications, it is more sensible to optimise certain compound performance metrics that aggregate over all the streams. Consequently, the decisions for different streams become dependent. We propose a general compound decision framework for parallel sequential change detection, under which different performance metrics are given. In addition, data-driven decision procedures are developed, and optimality results are established for them. Some simulation results will be given to show the power of the proposed method. This talk is based on the following papers, and several ongoing projects.
Chen, Y., Lee, Y-H, and Li, X. (2022). Item Quality Control in Educational Testing: Change Point Model, Compound Risk, and Sequential Detection. Journal of Educational and Behavioral Statistics. 47, 322–352.
Chen, Y. and Li, X. (2022+). Compound Online Changepoint Detection in Parallel Data Streams. Statistica Sinica. To appear in Volume 33, No. 1, 2023.
Lu, Z., Chen, Y. and Li, X. (2022+). Optimal Parallel Sequential Change Detection under Generalized Performance Measures. Submitted to IEEE Transactions on Signal Processing. Minor revision.
报告人简介：
Dr Yunxiao Chen is an assistant professor at London School of Economics and Political Science. His research focuses on the development of statistical and computational methods for solving problems in social and behavioural sciences, under three interrelated topics, including (1) large-scale item response data analysis, (2) measurement and predictive modelling based on dynamic behavioural data and (3) sequential design of dynamic systems with applications to educational assessment and learning. He has published in leading journals in psychometrics, statistics and machine learning, including Psychometrika, British Journal of Mathematical and Statistical Psychology, Journal of American Statistical Association, Biometrika, Annals of Applied Statistics, and Journal of Machine Learning Research. Dr Chen received the 2022 Psychometrics Society Early Career Award, 2018 NCME Brenda H. Lloyd Dissertation Award, and had been a United States National Academy of Education/Spencer Postdoctoral Fellow (2018-2020).
Before joining LSE, Dr Chen was an assistant professor in the Department of Psychology and the Institute for Quantitative Theory and Methods at Emory University. He completed his PhD in Statistics at Columbia University in 2016.