A Study on the ANN-Based Credit Risk Prediction Model and Its Application
Yan Ping Peng, Xuyan Tu · 2006
Credit risk early warning system must provide precise support information to the bank managers for decision-making. So the risk prediction model is critical and regarded as the core part of the whole risk early warning system. Compared with other traditional prediction models, the neural network model has the advantages of self-studying, self-organizing as well as self-adapting. This dissertation puts forward an ANN-based risk prediction model, which combines the Self-Organizing Map Neural Network and the Probabilistic Neural Network together. Furthermore, the improved iteration ways in constructing and training the model are also presented, which includes the SOM boundary effect processing and the rare samples handling. The established model is trained with financial ratios for a specific credit risk early warning experiment. The preliminary experimental result demonstrates that the SOM-PNN model performs better than some traditional ones in the rates of prediction accuracy and efficiency.