报告题目:Factor Analysis for Large-dimensional Matrix-valued Time Series
主 讲 人:张新生
报告时间:2021年12月13日14:00-15:00
报告地点:腾讯会议 会议 ID:432-683-280
点击链接入会: https://meeting.tencent.com/dm/obb2lyp5ZA8E
报告摘要:
In this talk, I will introduce a dimension reduction method for analyzing matrix-valued time series, that’s Matrix Factor Model (MFM). First I will review the projected estimation method for MFM, then I will give the rationale of the projection estimators of factor loadings and scores from the perspective of minimizing least squares objective function. Motivated by the least squares formulation, we further consider a robust method for estimating large-dimensional matrix factor model by utilizing Huber Loss function. We also propose an iterative procedure to estimate the pair of row and column factor numbers robustly.
主讲人介绍:
张新生教授简介:复旦大学统计学系系主任,博士生导师。现担任中国概率统计学会第十一届常务理事,曾任上海市数学会第十一届理事会常务理事、中国现场统计研究会生存分析分会副理事长、教育部高等学校数学与统计学教学指导委员会统计学专业教学指导分委员会委员。主要研究方向为:高维数据的统计推断、过程统计、随机过程及其应用等。在JRSSB、JMLR、JOE、中国科学等国内外权威期刊上发表学术论文60余篇。
邀请人:林路
欢迎各位老师同学积极参加!