会计信息的数据挖掘方法与银行信贷风险预测  

A Data Mining Approach to Extract Valuable Information in Accounting Data and a Model of Credit Risk Prediction

作　　者：彭寿康[1] 

机构地区：[1]浙江工商大学金融学院,浙江杭州310018

出　　处：《商业经济与管理》2008年第3期50-56,共7页Journal of Business Economics

基　　金：教育部人文社会科学规划课题(06JA790105)

摘　　要：基于信息熵理论和数据挖掘技术,文章提出一种会计信息的数据挖掘方法,投资者和债权人可用这种方法从企业的会计数据中挖掘出决策有用信息。本文以银行信贷风险预测模型构建为例,对这种方法的有效性进行了实证检验。Based on information entropy theory and data mining technologies, this paper proposes an approach to mining valuable hidden information in accounting data. Creditors and investors can use this approach to mine out valuable knowledge buried in accounting data. To demonstrate the effectiveness of this approach, an empirical example is given to show how we can use this approach to construct models of predicting credit risks. The results show that these models have achieved the prediction accuracy.

关 键 词：信贷风险 信息熵 数据挖掘 信号噪音差 

分 类 号：F830.9[经济管理—金融学]