News

STARF系列报告12月15日 | Liu Yanghui: Personalized treatment selection via the covariate - specific treatment effect curve for longitudinal data
发布时间:2022-12-12 

时间:2022年12月15日15:00-16:00

腾讯会议:918-115-552

主持人:汤银才教授

主办单位
( Statistical Theory and Related Fields 》编辑部
华东师范大学统计学院
华东师范大学统计交叉科学研究院
统计与数据科学前沿理论及应用教育部重点实验室

报告人简介
 Liu Yanghui , Doctor of Statistics , is a teacher at the School of Economics and Statistics , Guangzhou University . In March 2020, she graduated from East China Normal University majoring in Statistics . Her research interests contain functional and longitudinal data analysis , nonparametric and semi - parametric models . She has published several papers in journals such as the Journal of Econometrics , Medicine and Science in Sports and Ex - ercise , Journal of the Korean Society , etc .

报告摘要 
 Treatment selection based on patient characteristics has been widely recognized in modern medicine. In this paper , we propose a generalized partially linear single - index mixed - effects modeling strategy for treatment selection and heterogeneous treat - ment effect estimation in longitudinal clinical and observational studies . We model the treatment effect as an unknown functional curve of a weighted linear combina - tion of time - dependent COvariates . This method enables uS to investigate COvari - ate - specific treatment effects and make personalized treatment selection in a flexible fashion . We develop a method that combines local linear regression and penalized quasi - likelihood to estimate the weight for each covariate , the unknown treatment effect curve and the parameters for mixed effects . Based on pointwise confidence in - tervals for the treatment effect curve , we can make individualized treatment deci - sions from the information of patient characteristics .