李长平,张甜甜,宋德胜,胡良平.非配对设计多值名义资料多水平多重Logistic回归模型[J].四川精神卫生杂志,2019,32(6):495-500.
非配对设计多值名义资料多水平多重Logistic回归模型
Multilevel multiple Logistic regression analysis with the multi-value nominal data collected from the unpaired design
投稿时间:2019-11-19  
DOI:10.11886/scjsws20191119006
中文关键词:  多值名义资料  多水平  SAS软件  多重logistic回归分析
英文关键词:Multi-value nominal data  Multilevel  SAS software  Multiple logistic regression analysis
基金项目:国家高技术研究发展计划课题资助 2015AA020102国家高技术研究发展计划课题资助(2015AA020102)
作者单位E-mail
李长平 天津医科大学公共卫生学院卫生统计学教研室 1067181059@qq.com 
张甜甜 天津医科大学公共卫生学院卫生统计学教研室  
宋德胜 天津医科大学公共卫生学院卫生统计学教研室  
胡良平 世界中医药学会联合会临床科研统计学专业委员会  
摘要点击次数:
全文下载次数:
中文摘要:
      本文目的是介绍非配对设计多值名义资料多水平多重logistic回归模型的构建与求解方法。首先介绍了有关的基本概念,涉及“多值名义结果变量”“分层或多水平数据结构”和“扩展的多重logistic回归模型”;其次,呈现了一个具有二水平结构的横断面调查资料,该资料涉及多个影响因素和一个多值有序的结果变量(在本文中,将其视为多值名义结果变量);最后,借助SAS中的两个过程(即GLIMMIX和NLMIXED)对给定的资料进行统计分析,即构建和求解“非配对设计多值名义资料多水平多重logistic回归模型”,并对相关结果进行比较和解释。
英文摘要:
      The purpose of the paper was to introduce the construction and solution of multilevel multiple logistic regression analysis with the multi-value nominal data collected from the unpaired design. Firstly, the basic concepts related to “multi-value nominal outcome variables” “stratified or multilevel data structures” and “generalized multiple logistic regression models” were introduced. Secondly, a cross-sectional survey data with two-level was presented. The data contained many independent variables and a multi-value ordinal outcome variable (in this paper, it was treated as a multi-value nominal outcome variable). Finally, statistical analysis of data was performed by two procedures (GLIMMIX and NLMIXED) in the SAS software. Construction and solution of multilevel multiple logistic regression analysis with the multi-value nominal data collected from the unpaired design was preformed and the related output results were compared and explained.
      附件
查看全文  HTML  查看/发表评论  下载PDF阅读器
关闭
手机扫一扫