| 非配对设计多值名义资料一水平多重Logistic回归分析 |
| |
| 引用本文： | 巩晓文,李长平,胡良平.非配对设计多值名义资料一水平多重Logistic回归分析[J].四川精神卫生,2019,32(6):486-489. |
| |
| 作者姓名： | 巩晓文 李长平 胡良平 |
| |
| 作者单位： | 1.天津医科大学公共卫生学院卫生统计学教研室，天津 300070;2.世界中医药学会联合会临床科研统计学专业委员会，北京 100029;3.军事科学院研究生院，北京 100850 |
| |
| 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） |
| |
| 摘 要： | 本文目的是介绍非配对设计多值名义资料一水平多重logistic回归分析的基本原理、建模策略及注意事项。结合实例，应用SAS 9.4构建未经变量筛选和经变量筛选的多值名义资料多重logistic回归模型。通过回归分析的计算结果可知，同一变量的回归系数在不同logit函数中存在代数关系。多值名义多重logistic回归分析可以用来处理结果变量为多值名义变量的回归建模问题，并可以结合SAS实现对自变量的筛选，以获得简洁的回归模型。
|
| 关 键 词： | 多值名义资料 logistic回归分析 自变量筛选 设计矩阵 |
| 收稿时间： | 2019/11/19 0:00:00 |
One-level multiple Logistic regression analysis of the nominal data collected from the unpaired design |
| |
| Gong Xiaowen,Li Changping,Hu Liangping.One-level multiple Logistic regression analysis of the nominal data collected from the unpaired design[J].Sichuan Mental Health,2019,32(6):486-489. |
| |
| Authors: | Gong Xiaowen Li Changping Hu Liangping |
| |
| Institution: | 1.Department of Health Statistics, School of Public Health, Tianjin Medical University, Tianjin 300070, China;2.Specialty Committee of Clinical Scientific Research Statistics of World Federation of Chinese Medicine Societies, Beijing 100029, China;3.Graduate School, Academy of Military Sciences PLA China, Beijing 100850, China |
| |
| Abstract: | The paper introduced the basic principle, modeling strategy and key points of the one-level multiple logistic regression analysis of the multi-value nominal data collected from the unpaired design. In order to analyze the influencing factors of the choice of treatment for patients with non-ST-segment elevation myocardial infarction(NSTEMI) of the real example, and to predict the appropriate treatment according to the important characteristics, it was built that the one-level multiple nominal logistic regression model with and without variable selection by using SAS 9.4 software. The regression results showed that the regression coefficients of the same variable had algebraic relations in different logit functions. Multiple nominal logistic regression analysis could deal with the regression problem of the multi-value nominal data and with the help of SAS software. We could establish a concise model by filtering the insignificant independent variables. |
| |
| Keywords: | Multi-value nominal data Logistic regression analysis Variables selection Design matrix |
|
| 点击此处可从《四川精神卫生》浏览原始摘要信息 |
| 点击此处可从《四川精神卫生》下载免费的PDF全文 |
