合理进行多重Logistic回归分析——结合ROC曲线分析

Reasonably conduct the multiple Logistic regression analysis combined with the ROC curve analysis

本文目的是介绍如何结合ROC曲线分析,合理地进行多重Logistic回归分析的方法。第一,介绍了与ROC曲线分析有关的两组基本概念,即常用诊断指标的统计描述和诊断资料的ROC曲线分析方法。第二,介绍了ROC曲线分析中的核心内容,即ROC曲线下面积的计算和多条ROC曲线下面积的比较。第三,通过一个诊断试验的实例,介绍了如何用SAS软件进行分析的全过程,内容如下:①仅采用多重Logistic回归分析;②基于多重Logistic回归分析,再结合ROC曲线分析。得到的结论是,对于诊断试验资料,将多重Logistic回归分析与ROC曲线分析结合起来,可以获得更丰富、更合理的统计分析结果。

The purpose of this paper was to introduce how to reasonably carry out the method of the multiple Logistic regression analysis by combining the ROC curve analysis.Firstly,it introduced two groups of the basic concepts related to the ROC curve analysis,that was,the statistical description of common diagnostic indicators and the ROC curve analysis method of the diagnostic data.Secondly,it introduced the core contents of the ROC curve analysis,that was,the calculation of the area under the ROC curve and the comparison of the area under multiple ROC curves.Thirdly,through an example of a diagnostic test,the whole process of how to use SAS software for the analysis was introduced,the contents were as follows:①the analysis using only multiple Logistic regression analysis;②the multiple Logistic regression analysis combined with the ROC curve analysis.The conclusion was that,for the diagnostic test data,combining the multiple Logistic regression analysis with the ROC curve analysis could obtain richer and more reasonable statistical analysis results.