時間：2022年5月18日，星期三，上午

報告一(9:30---10:15)

題目：A New Projection Test for Mean Vector in High Dimensions

摘要：This paper studies the projection test for high-dimensional mean vectors via optimal projection. The idea of projection test is to project high-dimensional data onto a space of low dimension such that traditional methods can be applied. We propose a new estimation for the optimal projection direction by solving a constrained and regularized quadratic programming. The test is constructed using the estimated optimal projection direction. It is based on a data-splitting procedure, which achieves an exact t-test under normality assumption. To mitigate the power loss due to data-splitting, we further propose an online framework, which iteratively updates the estimation of projection direction when new observations arrive. Various simulation studies as well as a real data example show that the proposed online-style projection test retains the type I error rate well and is more powerful than other existing tests.

報告人簡介：鐘威，現(xiàn)任廈門大學(xué)王亞南經(jīng)濟研究院和經(jīng)濟學(xué)院統(tǒng)計系教授、博士生導(dǎo)師。2012年獲得美國賓夕法尼亞州立大學(xué)統(tǒng)計學(xué)博士學(xué)位，2014年和2017年分別破格晉升副教授和教授，2018年入選廈門大學(xué)“南強青年拔尖人才”（A類），國家自然科學(xué)基金優(yōu)秀青年基金獲得者（2019），福建省杰出青年基金獲得者（2019）。主要從事高維數(shù)據(jù)統(tǒng)計分析、統(tǒng)計學(xué)習(xí)算法、計量經(jīng)濟學(xué)、統(tǒng)計學(xué)和數(shù)據(jù)科學(xué)的應(yīng)用等研究。擔任美國統(tǒng)計協(xié)會(ASA)期刊《Statistical Analysis and Data Mining》和加拿大統(tǒng)計學(xué)會期刊《Canadian Journal of Statistics》的AE，在The Annals of Statistics, Journal of the American Statistical Association, Biometrika, Journal of Econometrics, Journal of Business & Economic Statistics, Biometrics, Annals of Applied Statistics, Statistica Sinica，中國科學(xué)數(shù)學(xué)等國內(nèi)外統(tǒng)計學(xué)權(quán)威期刊發(fā)表（含接收）20多篇論文，其中入選ESI前1%高被引論文2篇。主要講授《數(shù)理統(tǒng)計》、《廣義線性模型》、《計量經(jīng)濟學(xué)》、《統(tǒng)計數(shù)據(jù)分析》等本科和研究生課程，多次獲得學(xué)院教學(xué)優(yōu)秀獎，2016年獲得廈門大學(xué)第五屆英語教學(xué)比賽一等獎，2020年獲得廈門大學(xué)第十五屆青年教師技能比賽特等獎,2021年獲得廈門大學(xué)教學(xué)創(chuàng)新大賽一等獎，2021年獲得福建省“向上向善好青年”稱號。個人主頁：https://wise.xmu.edu.cn/people/faculty/bd5bc78c-99d3-46b0-873d-32fa79a278f5.html

報告二(10:20---11:05)

題目：A Generalized Knockoff Procedure for FDR Control in Structural Change Detection

摘要：Controlling false discovery rate (FDR) is crucial for variable selection, multiple testing, among other signal detection problems. In literature, there is certainly no shortage of FDR control strategies when selecting individual features. Yet lack of relevant work has been done regarding structural change detection, including, but not limited to change point identification, profile analysis for piecewise constant coefficients, and integration analysis with multiple data sources. In this paper, we propose a generalized knockoff procedure (GKnockoff) for FDR control under such problem settings. We prove that the GKnockoff possesses pairwise exchangeability, and is capable of controlling the exact FDR under finite sample sizes. We further explore GKnockoff under high dimensionality, by first introducing a new screening method to filter the high-dimensional potential structural changes. We adopt a data splitting technique to first reduce the dimensionality via screening and then conduct GKnockoff on the refined selection set. Numerical comparisons with other methods show the superior performance of GKnockoff, in terms of both FDR control and power. We also implement the proposed method to analyze a macroeconomic dataset for detecting change points in the consumer price index, as well as the unemployment rate.

報告人簡介：劉婧媛，廈門大學(xué)經(jīng)濟學(xué)院統(tǒng)計學(xué)與數(shù)據(jù)科學(xué)系、王亞南經(jīng)濟研究院教授、博士生導(dǎo)師，2021年入選國家級人才計劃。2013年博士畢業(yè)于美國賓夕法尼亞州立大學(xué)統(tǒng)計學(xué)專業(yè)?？蒲蟹矫嬷饕獜氖赂呔S數(shù)據(jù)分析、交叉學(xué)科的統(tǒng)計方法研究、統(tǒng)計基因?qū)W等領(lǐng)域的工作，在JASA，JOE, JBES, Annals of Applied Statistics等國際權(quán)威學(xué)術(shù)期刊發(fā)表論文20余篇；主持國家自然科學(xué)基金面上項目、青年項目等國家級、省部級多項科研項目；2018年入選福建省杰出青年科研人才培育計劃。教學(xué)方面曾獲廈門大學(xué)教學(xué)比賽特等獎、福建省一流課程等榮譽。

個人主頁：https://stats.xmu.edu.cn/info/1020/1055.htm

會議信息：

黃磊邀請您參加騰訊會議

會議主題：統(tǒng)計學(xué)系列學(xué)術(shù)報告

會議時間：2022/05/18 09:00-11:30 (GMT+08:00)中國標準時間-北京

點擊鏈接入會，或添加至?xí)h列表：

https://meeting.tencent.com/dm/e6yejbMVewe0

騰訊會議：884-226-111

會議密碼：220518