合理进行多重Logistic回归分析——结合平均处理效应分析

Reasonably conduct the multiple Logistic regression analysis combined with the average treatment effect analysis

中文关键词:  逆概率权重  潜在结果均值  平均处理效应  Logistic回归模型  倾向性评分模型

英文关键词:Inverse probability weight  Potential outcome mean  Average treatment effect  Logistic regression model  Propensity score model

中文摘要:

      本文目的是介绍如何结合平均处理效应分析，合理地进行多重Logistic回归分析的方法。第一，介绍了与平均处理效应分析有关的4个基本概念。第二，介绍了平均处理效应分析中的核心内容，即6种估算方法。第三，通过一个假设的药物临床试验实例，介绍了如何用SAS软件进行分析的全过程，内容如下：①采用传统的多重Logistic回归分析；②采用倾向性评分模型计算逆概率权重；③分别采用6种估算方法估计潜在结果均值和平均处理效应。

英文摘要:

      The purpose of the paper was to introduce how to reasonably carry out multiple Logistic regression analysis combined with the average treatment effect analysis. Firstly， it introduced 4 basic concepts related to the average treatment effect analysis. Secondly， it presented the core contents in the average treatment effect analysis， that was， six estimation methods. Thirdly， through a hypothetical drug clinical trial example， it gave the whole process of how to use SAS software for the analysis. The contests were as follows： ① the traditional multiple Logistic regression model was used for the analysis； ② the propensity score model was used to calculate the inverse probability weights； ③ six estimation methods were used to estimate the potential outcome mean and the average treatment effect.