合理进行均值比较——重复测量设计定量资料方差分析

Reasonably carry out mean value comparison:analysis of variance of the quantitative data of repeated measurement design

本文目的是介绍与混合模型有关的5个基本概念、计算方法、一个临床试验实例及其SAS实现。基本概念包括固定效应、随机效应、随机系数、混合模型和多水平模型。计算方法涉及混合模型的经典计算方法和贝叶斯计算方法。临床试验实例涉及三种药物对哮喘患者呼吸功能的影响。本文给出了采用SAS处理实例中定量资料的全过程,包括基于mixed过程和bglimm过程构建6个重复测量设计定量资料混合模型,对输出结果作出了解释,并基于模型拟合效果评价统计量,对所构建的多个混合模型进行了比较,得出了适合本例资料的最优混合模型。

The purpose of this article was to introduce 5 basic concepts,calculation methods,a clinical trial example and its SAS implementation related to the mixed model.The basic concepts include fixed effects,random effects,random coefficients,mixed models,and multilevel models.The calculation methods involved the classical calculation method and the Bayesian calculation method of the mixed model.An example of the clinical trial involved the effect of three drugs on the ability to breathe in people with asthma.This paper gave the whole process of using SAS software to implement the quantitative data in the example,including constructing 6 repeated measurement design quantitative data mixed models based on the mixed procedure and the bglimm procedure.The output results were explained,and all the mixed models constructed were compared based on the evaluation statistics of the model fitting effect,and the optimal mixed model suitable for the data in the paper was obtained.