报告题目：Symmetrized Data Aggregation for FDR Control

报告时间：2022年2月25日上午10：00

会议链接：https://meeting.tencent.com/dm/wjY15mSQGCDf

会议 ID：109-145-573

主办单位：数学与统计学院

主讲人：邹长亮

邹长亮简介：南开大学统计与数据科学学院教授。08年于南开大学获博士学位，随后留校任教。主要从事统计学及其与数据科学领域的交叉研究和实际应用。研究兴趣包括：高维数据统计推断、大规模数据流分析、变点和异常点检测等，在Ann.Stat.、Biometrika、 J.Am.Stat.Asso.、Math. Program.、Technometrics等统计学和工业工程领域期刊上发表论文几十篇，主持国家自然科学基金委项目多项。

报告摘要：We develop a new class of distribution–free multiple testing rules for FDR control. I will mainly illustrate the idea via multiple testing with general dependence. A key element in our proposal is a symmetrized data aggregation (SDA) approach to incorporating the dependence structure via sample splitting, data screening and information pooling. The SDA substantially outperforms the knockoff method in power under moderate to strong dependence, and is more robust than existing methods based on asymptotic p-values. I will also talk about some other applications, such as the selection of the number of change-points and threshold selection in feature screening.