摘要
本文依据商业银行信用风险的内涵,结合信用风险的不确定性和相对性特征,提出以"信用风险度"作为系统的输出,并针对传统模式识别评估方法的不足,构建了基于补偿模糊神经网络的信用风险评估预测模型,为有效转变信用风险的分类评估模式、提供更为全面的信贷决策支持奠定了基础。实证结果表明,该模型是一种较为有效的评估方法。
In view of the connotation of credit risk, credit risk degree is put forward by taking the probability that credit capital becomes bad debt and the relativity of credit risk into consideration. In allusion to the disadvantages of traditional pattern identification methods, credit risk assessing and forecasting model based on compensated fuzzy neural network is established so as to transform assessing mode in effect and provide credit decision-nnnmaking with more efficient tools and support. The empirical result of this paper suggests that this model is of relative effectiveness.
出处
《管理工程学报》
CSSCI
2007年第4期85-90,共6页
Journal of Industrial Engineering and Engineering Management
基金
国家自然科学基金资助项目(70373012)
关键词
信用风险评估
分类评估模式
补偿模糊神经网络
信用风险度
credit risk assessment
assessing mode by classification
compensated fuzzy neural network
credit risk degree