Study on Credit Risk Assessment Model of Commercial Banks Based on BP Neural Network.

An assessment model of commercial bank credit risk based on BP neural network is established, using the credit assessment index system established for commercial banks. This network is a 3-storey structure with 9 factor inputs and one output measuring the credit risk of commercial banks; besides, the neural network rule layer has the capability of self-adaptation, selforganization and self- adjustment in accordance with specific conditions or problems. After using the 3σ” Principle to assess the original credit risk of the enterprise, the standardized assessment indexes accord with the normal distribution and the value is 0.9974 in the section of 3σ” , and this method changes the situation using expert-marking method to ascertain the original credit level. A substantiation analysis has been applied to 161 groups of sample data using Matlab 7.0, and the training results indicate that the network prediction has less error, which has a great significance for commercial banks to reduce their credit risk.