报告题目：Calibration for non-positive definite covariance matrix
       主讲人：潘建新教授（英国曼彻斯特大学）
       时间：2019年6月24日（周一）14:30 p.m.
       地点：北院卓远楼305
       主办单位：统计与数学学院

       摘要：Covariance matrices that fail to be positive definite arise often in covariance estimation. Approaches addressing this issue exist, but are not well supported theoretically. In this paper, we propose a unified statistical and numerical matrix calibration method, finding the optimal positive definite surrogate in the sense of Frobenius norm. The proposed method is well supported theoretically and the proposed algorithm can be directly applied to any estimated covariance matrix. Numerical simulation results show that the calibrated matrix is typically closer to the true covariance, while making only limited changes to the original covariance structure. The proposed method is also applied to a real data analysis for illustration. This is a joint work with Chao Huang (University of Hull) and Daniel Farewell (Cardiff University).

       主讲人简介：
       潘建新，英国曼彻斯特大学数学学院统计系教授，威廉希尔体育,威廉体育官方：统计与数学学院特聘教授。于1996年在香港浸会大学获统计学博士学位。其研究方向包括纵向数据分析、生存数据分析、广义估计方程、生长曲线模型、均值与方差的同时建模，缺失数据问题及统计诊断。潘建新教授致力于统计方法的研究及其在生物、医学及金融领域内的应用研究。2002年在Springer出版社出版《生长曲线模型及统计诊断》专著（英文版；与方开泰教授合著）。目前担任Biometrics杂志编委（Associated Editor）。