The Warning Study Of Commercial Banks’ Liquidity Risk Based On BP Neural Network

Analyzing the effect of the forewarning of the commercial bank’s liquidity risk isvery important for its development. Many scholars studied the problem of themanagement of the commercial bank’s liquidity risk, but they did not reach aconsensus. It is noteworthy that previous studies were carried out the qualitativeanalysis of this problem. This paper will use the BP neural network,GM(1,1) modeland Principal component analysis combining to the study of the forewarning of thecommercial bank’s liquidity risk.Chapter1introduces the background and significance of this thesis, literaturereview, thesis structure and research methods. Chapter2is the part of the technicaldesign. This chapter mainly discusses the neural network, the BP neural network,reviews the neural network theory,and focus on the BP neural network’s correlationproperties and application. Lastly,this chapter introduces Principal componentanalysis theory. Chapter3analyzes the causes and measurement of the commercialbank liquidity risk.Then elaborates the construction of index system. Chapter4is thecore part.According to the content of Chapter2and3, this article selects a sample ofthe commercial banks,and then do the empirical analysis of the correspondingdata.The result of Principal component analysis reveales that numerous indexes havesome relevances.We can select some common factors to replace the similarindexes.Then according to the results of principal component analysis, we define therisk level of this sample.We use the BP neural network and GM(1,1) model tosimulate the sample data,and the results show the accuracy rate of the BP neuralnetwork is better,reaching82%.This shows that the BP neural network model can bebetter applied to the forewarning of the commercial bank’s liquidity risk,and it isfeasibility and rationality.Lastly, according to the results of empirical analysis onbank liquidity risk,the author puts forward some corresponding countermeasures. Theconclusion part summarizes the paper.This paper uses BP neural network model to study the forewarning of thecommercial bank’s liquidity risk and carry out empirical analysis, which has strong theoretical and practical significance. The conclusion of this paper has certainreference function.