报告题目：
Robust Estimation for Longitudinal Data with Covariate Measurement Errors and Outliers
针对协变量带有测量误差和异常值的纵向数据的稳健估计
报告时间：2019年4月16日（星期二）上午10:30-11:30
报告地点：开云·电竞(中国)官方网站二楼会议室
报告人：秦国友，复旦大学公共卫生学院，教授，博士，硕士生导师。
研究方向包括纵向数据分析，半参数模型的稳健推断，缺失数据分析以及统计方法在公共卫生和医学中的应用，主要关注肿瘤，慢性病领域的统计方法应用研究。
在国际上，主要和香港大学、美国伊利诺伊大学厄巴纳－香槟分校、美国北卡罗来纳大学教堂山分校生物统计系、新加坡南洋理工大学物理与数学科学学院等合作研究。在SCI源期刊上发表几十篇高质量论文，包括一些在国际高引著名杂志Biometrics, Biostatistics上发表的论文。 2014年度教育部高等学校科学研究优秀成果奖二等奖(第二完成人)，2015年 入选复旦大学“卓学计划”，担任中华预防医学会生物统计专业青年委员会主任委员，中华预防医学会生物统计专业委员会 委员，中国卫生信息学会统计理论与方法专业委员会委员，担任《中国卫生统计》,《复旦学报（医学版）》编委，《美国数学评论》评论员。主持多项国家自然基金和省部级项目基金。
报告摘要：
Measurement errors and outliers often arise in longitudinal data, ignoring the effects of measurement errors and outliers will lead to seriously biased estimators. Therefore, it is important to take them into account in longitudinal data analysis. In this paper, we develop a robust estimating equation method for analysis of longitudinal data with covariate measurement errors and outliers. Specifically, we eliminate the effects of measurement errors by making use of the independence of replicate measurement errors and correct the bias induced by outliers through centralizing the matrix of error-prone covariates in the estimating equation. The proposed method is easy to implement by using the standard generalized estimating equations algorithms and does not require specifying the distributions of the true covariates, response and measurement error. The asymptotic normality of the proposed estimator is established under some regularity conditions. Extensive simulation studies show that the proposed method does have a good performance in handling measurement errors and outliers. In the end, the proposed method is applied to data from the Lifestyle Education for Activity and Nutrition (LEAN) study for illustration.