异质性视角下Meta分析的统计诊断研究  

Statistical Diagnosis of Meta Analysis From the Perspective of Heterogeneity

作　　者：张敏 李勇 Zhang Min;Li Yong(Research Center for Economy of Upper Reaches of the Yangtze River,Chongqing Technology and Business University,Chongqing 400067,China;School of Statistics,Chengdu University of Information Technology,Chengdu 610103,China)

机构地区：[1]重庆工商大学长江上游经济研究中心,重庆400067 [2]成都信息工程大学统计学院,成都610103

出　　处：《统计与决策》2021年第21期41-45,共5页Statistics & Decision

基　　金：重庆市科学技术研究项目(KJQN202100806);重庆市社会科学规划项目(2020QNJY59,2020ZDTJ08);农业现代化与产业创新发展研究团队资助项目(CJSYTD201710);重庆市教育科学“十三五”规划课题(2020-GX-294)。

摘　　要：Meta分析是将同一问题的多个独立研究结果汇总,利用统计建模进行定量综合分析的一种系统评价方法。文章基于异质性视角研究了随机效应Meta回归数据删除法、替换法和均值漂移的统计诊断模型等价性及其条件。构造其变化统计量和Cook距离统计量,并比较验证了一个实际例子关于三种诊断方法的多个异常值或强影响点的识别效果。Meta analysis is a systematic evaluation method that summarizes the results of multiple independent studies on the same issue and makes quantitative comprehensive analysis by statistical modeling. This paper studies the equivalence and conditions of statistical diagnostic models for data deletion, substitution and mean shift of random effects Meta regression from the perspective of heterogeneity. And then, the paper constructs the changing statistics and Cook distance statistics, and also compares a practical example to verify the identification effect of multiple outliers or strong influence points of three diagnostic methods.

关 键 词：META分析 异质性 统计诊断 

分 类 号：O212[理学—概率论与数理统计] F224[理学—数学]