報告人：李忠華 教授

報告題目：A Unified Diagnostic Framework via Symmetrized Data Aggregation

報告時間：2025年11月7日（周五）14:00-15:00

報告地點：云龍校區(qū)6號樓304會議室

主辦單位：數(shù)學與統(tǒng)計學院、數(shù)學研究院、科學技術研究院

報告人簡介：

李忠華，南開大學統(tǒng)計與數(shù)據(jù)科學學院教授，曾受邀訪問美國北卡羅萊納大學教堂山分校、明尼蘇達大學等。研究方向為統(tǒng)計質(zhì)量控制、變點、高維統(tǒng)計推斷、網(wǎng)絡數(shù)據(jù)分析等。合作出版專著1本，發(fā)表學術論文50余篇。現(xiàn)任中國數(shù)學會概率統(tǒng)計分會副秘書長、中國現(xiàn)場統(tǒng)計研究會統(tǒng)計學歷史與文化分會副理事長、中國優(yōu)選法統(tǒng)籌法及經(jīng)濟數(shù)學學會工業(yè)工程分會常務理事、全國工業(yè)統(tǒng)計學教學研究會理事、國際質(zhì)量工程期刊Quality Engineering編委、美國Mathematical Reviews評論員等。

報告摘要：

In statistical process control (SPC) of high-dimensional data streams, besides online monitoring of abnormal changes, fault diagnosis of responsible components has become increasingly important. In this talk, we introduce a new procedure to control the false discovery rate (FDR) of fault diagnosis. The proposed method formulates the fault diagnosis as a variable selection problem and utilizes the symmetrized data aggregation (SDA) technique via sample splitting, data screening, and information pooling to control the FDR. Under some mild conditions, we show that the proposed method can achieve FDR control asymptotically. Extensive numerical studies and real-data examples demonstrate satisfactory FDR control and remarkable diagnostic power in comparison to existing methods.