基于广义病例-队列设计方案的长度偏差数据回归分析

吉林大学学报(理学版) ›› 2019, Vol. 57 ›› Issue (2): 311-316.

基于广义病例-队列设计方案的长度偏差数据回归分析

徐达¹, 周勇^2,3,4

1. 上海财经大学统计与管理学院, 上海 200433; 2. 华东师范大学统计交叉科学研究院, 上海 200241;3. 华东师范大学统计学院, 上海 200241; 4. 中国科学院数学与系统科学研究院, 北京 100190

收稿日期:2018-07-16 出版日期:2019-03-26 发布日期:2019-03-26
通讯作者: 徐达 E-mail:xuda369@foxmail.com

Regression Analysis for LengthBiased Data Based on Generalized CaseCohort Design Scheme#br#

XU Da^1, ZHOU Yong^2,3,4

1. School of Statistics and Management, Shanghai University of Finance and Economics, Shanghai 200433, China;2. Institute of Statistics and Interdisciplinary Sciences, East China Normal University, Shanghai 200241, China;3. School of Statistics, East China Normal University, Shanghai 200241, China;4. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China

Received:2018-07-16 Online:2019-03-26 Published:2019-03-26
Contact: XU Da E-mail:xuda369@foxmail.com

摘要/Abstract

摘要： 利用加权估计方程方法, 在广义病例队列设计方案下, 针对长度偏差数据, 给出Cox模型中回归系数的估计, 并证明在适当的条件下, 所得估计量具有相合性和渐近正态性, 且渐近方差具有显式表达, 在实际应用中可由plugin方法估计.

关键词: 长度偏差数据, 广义病例队列设计, Cox比例风险模型, 混合估计方程

Abstract: Using the method of weighted estimation equation, under the generalized casecohort design scheme, we gave the estimation of regression coefficients in Cox’s model for lengthbiased data. Under appropriate conditions, we proved that the obtained estimators were consistent and asymptotic normality, and the asymptotic variance was explicit expression. It could be estimated by the plugin method in practical application.

Key words: lengthbiased data, generalized casecohort design, Cox proportional risk model, composite estimation equation

中图分类号:

O212.1

徐达, 周勇. 基于广义病例-队列设计方案的长度偏差数据回归分析[J]. 吉林大学学报(理学版), 2019, 57(2): 311-316.

XU Da, ZHOU Yong. Regression Analysis for LengthBiased Data Based on Generalized CaseCohort Design Scheme#br#[J]. Journal of Jilin University Science Edition, 2019, 57(2): 311-316.