详细信息

基于混合系统的信用风险评估  ( EI收录)  

Credit risk evaluation based on hybrid system

文献类型:期刊文献

中文题名:基于混合系统的信用风险评估

英文题名:Credit risk evaluation based on hybrid system

作者:马海英[1]

机构:[1]华东理工大学管理科学与工程系,上海200237

年份:2006

卷号:46

期号:Z1

起止页码:1099

中文期刊名:清华大学学报(自然科学版)

外文期刊名:Journal of Tsinghua University(Science and Technology)

收录:CSTPCD;;EI(收录号:20070110348823);Scopus;北大核心:【北大核心2004】;CSCD:【CSCD2011_2012】;

语种:中文

中文关键词:信用风险;人工智能;神经网络;专家系统

外文关键词:credit risk;artificial intelligence(AI);neural network;expert system

摘要:提出一种基于混合系统的信用风险评估方法,用于识别和评价中国商业银行信用风险。用自适应共振神经网络模型进行风险的定量分析,用专家系统进行定性分析,结合定量分析的结果,给出分析结论。实证分析的结果表明,对于统计方法和BP模型而言,自适应共振模型的误判率低,且风险分类的精度高,从而提高了整个混合系统评估的准确性。该方法具有较强的可操作性,可以得到较好的评估效果,适合于于中国的信用风险数据基础薄弱的情况。
This paper presented a credit risk evaluation method based on hybrid system to establish the identification and the evaluation credit risk system of China Commercial Bank.The ART2 model was used to make quantitative analysis.Expert system was used to have qualitative analysis.The ART2 has the lower misjudging rate compared with statistical method and BP model,which can increase the accuracy of whole hybrid system.The results show that this method has strong operational characters in the credit risk evaluation,and is effective and suitable for the foundation data weakness condition in China.

参考文献:

正在载入数据...

版权所有©华东理工大学 重庆维普资讯有限公司 渝B2-20050021-7 
渝公网安备 50019002500408号 违法和不良信息举报中心