Analysis of Bank Loan Risk Management Based on BP Neural Network

Abstract

Credit risk in financial institutions, especially banking, has a long history and is a universal problem in the world. As far as our country is concerned, the credit risk management of commercial banks is not perfect, and the theory and technology are relatively simple, so it is far from solving all kinds of situations in the risk management of commercial banks in our country. In this paper, the selection range of an index is given under the condition of following the selection principle. On the basis of analyzing the basic characteristics of neural network and artificial neural network, this paper focuses on the classification method of loan risk based on three-layer single-node output BP neural network (BPNN), which divides listed enterprises into normal, concerned, secondary, suspicious and loss from the perspective of loan risk. The results are tested, and the superiority of BPNN in overall risk assessment is proved by the test of Logistic model.