基于RBF网络的商业银行信用风险控制研究

对信用风险的有效控制与管理,在现代商业银行日常运行过程中具有举足轻重的地位。基于信用风险系统是一个高度复杂的非线性动态系统,利用神经网络的自适应学习、并行分布处理和较强的鲁棒性及容错性等特性,建立基于RBF神经网络的信用风险预测控制模型,从理论上探寻信用风险非线性智能控制。仿真试验表明,信用风险度能被控制在以最佳风险度为中心的一定范围内。因此,该预测控制系统适合于商业银行信用风险的控制。

A Study of Credit Risk Control by Commercial Banks Based on a RBF Artificial Neural Network

It is highly important for modern commercial banks to effectively control credit risks. Credit risk system is a very complicated non-linear dynamic system. A credit risk forecast and control model based on RBF neural network with the characteristics of adaptive learning, parallel distribution processing, strong robustness and fault tolerance has been found and used to make a theoretical probe into non-linear intelligence control of credit risks. Simulation indicates that credit risks can be limited within an optimum risk range. So, the forecast and control model is applicable to commercial banks for credit risk control.