Speaker: Prof. LI Qizhai, University of Chinese Academy of ScienceVenue: Tencent meeting, meeting ID: 559 530 381Abstract:Modern high-dimensional statistical inference often faces the problem of missing data. In recent decades, many studies have focused on this topic and provided strategies including complete-sample analysis and imputation procedures. However, complete-sample analysis discards information of incomplete samples, while imputation procedures have accumulative errors from each single imputation. In this paper, we propose a new method, Sample-wise COmbined missing effect Models with penalization (SCOM), to deal with missing data occurring in predictors. Instead of imputing the predictors, SCOM estimates the combined effect caused by all missing data for each incomplete sample. SCOM makes full use of all available data and is robust with respect to various missing mechanisms. Theoretical studies show the oracle inequality for the proposed estimator, and the consistency of variable selection and combined missing effect selection. Simulation studies and an application to the Residential Building Data also illustrate its effectiveness.

Speaker: FU Chengde, professor of Fanhai International School of Finance, Fudan UniversityVenue: Room 200-9, Run Run Shaw Business Administration Building, Yuquan CampusAbstract:Existing empirical studies on correlated defaults have shown that the default of a firm impacts other firms; however, this impact has yet to be theoretically validated and quantified, especially under a structural-form model with more than two firms, and with multiple firms that are all likely to default with almost equal weight. To fill the gap, we study how the firm value processes interact with each other in the presence of correlated defaults as well as a large number of firms. To this end, a new renewal theory is developed. The results show that even under a simple one-factor model, the idiosyncratic moments of the defaulted firm transfer to other firms at the time of default, causing a propagation in credit risk. Furthermore, we can quantify this propagation via asymptotic theory, which provides a multiname distanceto-default type risk measure for a system of firms. The results potentially constitute a new method in studying contagion and other correlated default effects and therefore provide new measurements in credit risk management. Numerical and empirical studies are presented to illustrate our claim. 

Speaker: LIU Rui, professor, department of pure mathematics, Nankai UniversityVenue: Tencent meeting, meeting ID: 213 232 241Constructing dilations of frame decompositions and (non-commutative) operator-valued measures on Banach spaces.pdf

Speaker: Lucio Soibelman, chair man of Massachusetts Institute of Technology, Thouthern California University

Speaker: DENG Jintao， Postdoctoral Fellow, Department of Mathematics, University of Waterloo, CanadaVenue: Tencent meeting, meeting ID: 978 479 233The K-theory of Roe algebras and the coarse Baum-Connes conjecture.pdf

Speaker: DING Jian, professor, University of PennsylvaniaVenue: Room 1417, Administration Building， Zjingang Campus, Zhejiang University

Speaker: YANG Beijian, senior lecturer of Microsoft Office series coursesVenue：Library Multi-function Room, 1st Floor, International Campus

Speaker: SHOU Huixia, Professor, college of life science, Zhejiang UniversityVenue：Faculty Club #4, International Campus

Speaker:CHEN Zhenqing, professor, department of mathematics, the University of Washington.Venue: Tencent meeting, meeting ID: 273 754 387