報(bào) 告 人：蔣杭進(jìn) 研究員

報(bào)告題目：Two-sample Distribution Comparison and Its Applications 

報(bào)告時(shí)間：2023年07月24日（周一）上午10：00

報(bào)告地點(diǎn)：靜遠(yuǎn)樓1506會(huì)議室

主辦單位：數(shù)學(xué)與統(tǒng)計(jì)學(xué)院、數(shù)學(xué)研究院、科學(xué)技術(shù)研究院

報(bào)告人簡(jiǎn)介：

       蔣杭進(jìn)，香港中文大學(xué)統(tǒng)計(jì)學(xué)博士，現(xiàn)任浙江大學(xué)“百人計(jì)劃”研究員，博導(dǎo)，浙江省生物信息學(xué)學(xué)會(huì)理事、副秘書長(zhǎng)。2022年入選省級(jí)青年人才計(jì)劃。2023年成為教育部生物學(xué)科”101計(jì)劃”生物信息學(xué)核心課程建設(shè)專家。他曾訪問香港理工大學(xué)應(yīng)用數(shù)學(xué)系和斯坦福大學(xué)統(tǒng)計(jì)系Wing Hung Wong 教授。目前, 他主要致力于以統(tǒng)計(jì)方法推動(dòng)生命科學(xué)的發(fā)展, 為分析生物醫(yī)學(xué)數(shù)據(jù)提出新方法。研究興趣為貝葉斯數(shù)據(jù)分析、生物統(tǒng)計(jì)/計(jì)算生物學(xué)、統(tǒng)計(jì)推斷和深度學(xué)習(xí)。目前，他已經(jīng)在Nature Astronomy, The Astrophysical Journal, Statistica Sinica, Methods, PLoS Computational Biology, Briefings in Bioinformatics, IEEE-TCBB, The Plant Journal,和Advanced Biology等生物學(xué)、統(tǒng)計(jì)學(xué)和天文學(xué)領(lǐng)域的高水平期刊上發(fā)表論文20余篇。主持國(guó)家自然科學(xué)基金青年項(xiàng)目、浙江省人民醫(yī)院委托項(xiàng)目一項(xiàng)和浙江省農(nóng)科院委托項(xiàng)目一項(xiàng)，參與國(guó)家自然科學(xué)基金面上項(xiàng)目多項(xiàng)、浙江省重大研發(fā)計(jì)劃( 排名第2)一項(xiàng)和中國(guó)煙草總公司重大科技計(jì)劃(子課題負(fù)責(zé)人)。

報(bào)告摘要：

       We proposed nonparametric test statistics for the problem of the two-sample distribution comparison. These test statistics combine the merits of the chi-squared and Kolmogorov–Smirnov statistics, and provide new insights into the equality test of the unspecified distributions underlying the two independent samples. Based on our new statistics, we propose a marginal screening procedure and a pairwise joint screening procedure for detecting important variables in high-dimensional binary classification. Both screening procedures have the consistent screening property, which is stronger than the sure screening property of most existing methods. The marginal screening procedure is much more powerful than other methods over a broad range of cases, and the pairwise joint screening procedure provides a way of detecting variables with a joint effect, but no marginal effect. Extensive simulations and a real-data application show the effectiveness and advantages of the proposed methods.