光华讲坛——社会名流与企业家论坛第 期
主 题:Inference on Potentially Identified Subgroups in Clinical Trials临床试验中潜在识别亚组的推断
主讲人:香港科技大学郭心舟博士
主持人:统计学院陈雪蓉教授
时间:2024年5月24日(周五)下午2:30-3:30
举办地点:柳林校区弘远楼408会议室
主办单位:统计研究中心和统计学院 科研处
主讲人简介:
Xinzhou Guo is an Assistant Professor in the Department of Mathematics at the Hong Kong University of Science and Technology. He received his B.S. in Applied Mathematics from Peking University and Ph.D. in Statistics from the University of Michigan. Prior to joining HKUST in 2021, he did a postdoc at Harvard University. His main research interests are subgroup analysis, resampling methods, precision medicine and regulatory decision-making.
郭心舟是香港科技大学数学系的助理教授。他获得了北京大学应用数学学士学位和密歇根大学统计学博士学位。在2021年加入香港科技大学之前,他在哈佛大学进行了博士后研究。他的主要研究兴趣包括亚组分析、重抽样方法、精准医学和监管决策。
内容简介:
When subgroup analyses are conducted in clinical trials with moderate or high dimensional covariates, we often need to identify candidate subgroups from the data and evaluate the potentially identified subgroups in a replicable way. The classical statistical inference applied to the potentially identified subgroups, assuming the subgroups are the same as what we observe from the data, might suffer from bias issue when the regularity assumption that the boundaries of the subgroups are negligible is violated. In this paper, we propose a shift-based method to address nonregularity bias issue and combining it with cross-fitting and subsampling, develop a de-biased inference procedure for potentially identified subgroups. The proposed method is model-free and asymptotically efficient whenever it is possible, and can be viewed as an asymmetric smoothing approach. The merits of the proposed method are demonstrated by re-analyzing the ACTG 175 trial. This talk is based on joint work with Shuoxun Xu (HKUST).
在具有中高维协变量的临床试验中进行亚组分析时,通常需要从数据中识别候选亚组,并以可重复的方式评估潜在识别的亚组。经典的统计推断应用于潜在识别亚组时,假设这些亚组与我们从数据中观察到的一致,但当亚组边界的规则性假设被违反时,可能会出现偏差问题。本文提出了一种基于偏移的方法来解决非规则性偏差问题,并结合交叉拟合和子抽样,开发出一种去偏推断程序,用于潜在识别亚组。所提出的方法是无模型的,并在可能的情况下渐近有效,可以看作是一种非对称平滑方法。通过重新分析ACTG 175试验,展示了该方法的优点。本报告基于与Shuoxun Xu(香港科技大学)的合作研究。