时间：2019年6月14日(周五)上午10:00-11:00
地点：闵行校区法商南楼135室
题目：Empirical likelihood inference for longitudinal data with covariate measurement errors: an application to LEAN study
报告人：秦国友 教授 复旦大学
摘要：Measurement errors usually arise during the longitudinal data collection process and ignoring the effects of measurement errors will lead to invalid estimates. The Lifestyle Education for Activity and Nutrition (LEAN) study (Barry et al., 2011) was designed to assess the effectiveness of intervention for enhancing weight loss over a 9-month period in sedentary overweight or obese adults. The covariates systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured at baseline, month 4 and month 9. At each assessment time, there were two replicate measurements for SBP and DBP, where the replicate measurement errors of SBP and DBP respectively followed different distributions. In order to account for the different distributions of measurement errors, we develop a new method for analysis of longitudinal data with replicate covariate measurement errors based on the empirical likelihood method. The asymptotic properties of the proposed estimator are established under some regularity conditions and the confidence region for the parameters of interest can be constructed based on the chi-squared approximation without estimating the covariance matrix. Additionally, the proposed empirical likelihood estimator is asymptotically more efficient than the estimator of Lin et al. (2018). Extensive simulations also demonstrate that the proposed method can eliminate the effects of measurement errors in the covariate and has a high estimation efficiency. The proposed method indicates the significant effects of intervention, SBP and assessment time on BMI in the LEAN study.
报告人简介：秦国友，复旦大学公共卫生学院生物统计系系主任，博士，教授，硕士生导师，曾于2008.9-2009.9在美国北卡罗来纳大学教堂山分校生物统计系进行博士后研究。主要研究方向为：纵向数据分析、缺失数据分析、测量误差数据分析、分位数回归、稳健推断、部分线性模型和复杂抽样设计，主持多项国家级和省部级科研项目，在国际重要统计学期刊上发表SCI学术论文40多篇，其中多篇论文发表在国际顶级统计学杂志：Biometrics、Biostatistics上。