报告题目：UNCONDITIONAL QUANTILE REGRESSION FOR STREAMING DATA: A Modern Regression Model In  Big Data

报告时间：2025-07-01   10:00-11:00

报  告 人 ：虞克明  教授  （伦敦布鲁内尔大学）

报告地点：雷军科技楼六楼报告厅（601）

Abstract:This talk starts from a brief review of statistical analysis of big data, then introduces one of the modern regression models, namely Unconditional Quantile Regression (UQR),  which was initially introduced by Firpo et al. (2019) and has  gained significant traction as a popular approach for modelling and analyzing data. However, much like Conditional Quantile Regression (CQR), UQR encounters computational challenges when it comes to obtaining parameter estimates for big data (streaming datasets). This is attributed to the involvement of unknown parameters in the logistic regression loss function used in UQR, which presents obstacles in both computational execution and theoretical development. To address this, we present a novel approach involving smoothing logistic regression estimation. Subsequently, we propose a renewable estimator tailored for UQR with streaming data, relying exclusively on current data and summary statistics derived from historical data.