专题链接

学术看板 当前位置:网站首页 > 专题链接 > 学术看板

学术讲座:美国罗格斯大学谢敏革教授讲座

作者:大数据研究中心 来源:大数据研究中心 阅读次数:336日期:2018/05/15

标题: Uncertainty Quantification of Treatment Regime in Precision Medicine by Confidence Distributions

时间:5月20日19:30-20:30

地点:创新中心311会议室

报告人:谢敏革   教授

Abstract: Personalized decision rule in precision medicine can be viewed as a dis-crete parameter, for which theoretical development for statistical inference is lagged behind. In this talk, we propose a new way to quantify the estima-tion uncertainty in a personalized decision based on recent developments of confidence distribution (CD). Specifically, in a parametric regression model setup, suppose the binary decision for treatment versus control for an in-dividual xa is determined by a linear decision rule Da = 1(xaβ > xaγ), where β and γ are unknown regression coefficients in models for potential outcomes of treatment and control, respectively. The data-driven decision Da = 1(xaβˆ > xaγˆ) relies on the estimates βˆ and γˆ, which in turn intro-duces uncertainty on the decision. In this work, we propose to find a CD for ηa = xaβ − xaγ and compute a confidence measure of {Da = 1} = {ηa > 0}. This measure has a value between 0 and 1, and provides a frequency-based assessment on how reliable our decision is. For example, if the confidence measure of the decision {Da = 1} is 63%, then we know that, out of 100 pa-tients who are the same as patient xa, 63 will benefit to have the treatment and 37 will be better off to be in the control group. This new confidence mea-sure is inline with the classical assessments of sensitivity and specificity; but different from the classical assessments, this measure can be directly com-puted from the observed data without the need to know the truth whether {Da = 1} or {Da = 0}. Utility of this new measure will be illustrated in an application to design adaptive clinical trials.

Short Bio: Dr. Min-ge Xie is Distinguished Professor of Statistics from Depart-ment of Statistics and Biostatistics, Rutgers University. His main research interest lies in developing new statistical methodologies and theories for problems stemming from interdisciplinary research. His research interests include confidence distribution and foundation of statistical inferences, fu-sion learning, meta-analysis, estimating equations, asymptotic theories, and applications in bio-medical sciences, social sciences, industry, and engineer-ing. His research has been funded in part by grants from the National Science Foundation (NSF), the National Institute of Health (NIH), the Department of Homeland Security (DHS), the Federal Aviation Administration (FAA), among others.