报告题目：Covariance estimation in multidimensional response linear mixed effects models

报告人：王跃东 教授

报告时间：2026年7月20日10:00-11:00

报告地点：数学与统计学院210会议室

邀请人：曾佑杰

邀请单位：福州大学数学与统计学院

报告内容简介：We introduce a flexible moment-based method for estimating covariance matrices in multidimensional response linear mixed effects models (MLMM). Our primary methodology entails formulating individual nonlinear semidefinite programming problems for estimating each covariance matrix, thereby establishing an efficient solution framework. Compared to existing likelihood-based and Bayesian approaches, our method is several orders of magnitude faster and capable of handling moderately large response dimensions. We established non-asymptotic upper bounds on the estimation error for the proposed estimators, which are also verified in simulation studies. We apply this method to analyze clinical and laboratory variables from hemodialysis patients, examining their trajectories over time and uncovering networks at both the patient and treatment levels.

报告人简介：王跃东，美国加州大学圣巴巴拉分校统计与应用概率系讲席教授，是统计学界具有卓越贡献的研究者，为国际统计学会当选会士、美国统计协会当选会士、英国皇家学会会士，是国际数理统计协会、泛华统计协会、国际统计科学学会的会员。研究领域包括平滑样条、混合效应模型、生存分析、纵向数据、微阵列数据分析等方向，并在统计学国际顶尖学术期刊（Journal of the American Statistical Association、Annals of Statistics、Journal of the Royal Statistical Society、Biometrika 等）发表高水平论文三十余篇。其专著《Smoothing Splines: Methods and Applications》于2011年由Chapman & Hall/CRC出版。