基于BP算法的信用风险评价模型研究

Credit-Risk Evaluation Model Based on Back-Propagation Algorithm

本文利用神经网络技术建立基于 BP算法的信用风险评价模型 ,为我国某商业银行 12 0家贷款企业进行信用风险评价 ,按照企业的信用等级分为“信用好”、“信用中等”和“信用差”三个小组 .仿真结果表明 ,本文所建立的神经网络信用风险评价模型的分类准确率高于传统的参数统计分类方法——线性判别分析法的分类准确率 .文中还详细给出神经网络信用风险评价模型的网络构建方法及基于 BP网络的学习算法和步骤 .

The research uses neural network tec hnology to establish a credit-risk evaluation model based on back-propagation algorithm. The model is used to evaluate credit risk for 120 applicants of a co mmercial bank in our country. According to their credit grades, the applicants a re separated three groups: a ″good credit″ group, a ″middle credit″ group an d a ″bad credit″ group. The simulation shows that the neural network credit-risk evaluation model has higher classification accuracy compare with a tradit ional parameter statistical approach, that is linear discriminant analysis. We s till introduce the constructed method of the network in detail and give a learni ng algorithm and steps based on back-propagation algorithm.