递归分类树在信用风险分析中的应用

递归分类树是计算机实现 ,基于统计理论的非参数的识别技术 .本文将之用于商业银行信用风险分析 ,以判别分析为参照方法 ,加以实证研究 .结果表明递归分类树在变量选择和分类方面的效果好于判别分析方法.

Application of Classification Tree in Credit Risk Analysis

Classification tree shows promising application in credit risk analysis as a statistic\|based pattern recognition computer\|technique. The paper presents the necessity to extend the study of classification tree in credit risk analysis after comparing it with methods of discriminant analysis. Studies are empirically carried out, and prefer classification tree to statistical methods in classification accuracy and variable selection.