Commercial Bank Credit Risk Early Warning System Based On Bp Neural Network And Expert System Research

Since the 1970s, the trend of financial liberty and globalization have developed quickly, Financial innovation emerged in endlessly. The risk conditions of bank come to complex increasingly, banking supervision measurement and credit risk management was not available to it. People paid the enormous attention to the crisis that poor commercial bank credit risk management resulted in. It enriched greatly the ideas and technology of commercial bank credit risk management, promoted 'New Basel Capital Accord' which emphasized particularly on bank credit risk management to promulgate. It brought credit risk management to be the chief strategic issue. The study of 'world bank' about global banking crisis shows that the poorness of credit risk management is the common reason to result in commercial bank collapse. Recently Academe and Finance industry pay abroad attention to the commercial bank credit risk. With the development of the financial liberty and integration, especially China acceding to WTO, Chinese commercial bank will be confronted with severe circumstance in the course of financial revolution, which will bring about the commercial bank's risk more and more. Of course, it is very necessary to research and grasp completely and systematically the type, characteristic, manifest and reason, and construct the early warning system of the commercial bank's risk to control completely the risk。Artificial neural network is the imitation on the basis of the man's brain. Whether in thoughts or technology, application of the ANN in early warning of commercial bank credit risk is a breakthrough. Commercial bank credit risk management can be divided into two sections, early warning before loan and early warning after loan. The purpose of this paper is to study the early warning after loan. At the beginning of this paper, we discuss the early warning signal about finance factors , introduce the back-propagation neural network to design network framework, carry on the network training and test, then make the positive analysis to the samples with Matlab. According to the results that B-P model analyze, carry on the general analysis with the expert system, thus achieve the goal of credit loan early warning. At last, we discuss the problems that exist in the commercial bank...