Research on financial risk early warning based on nonparametric test combined with deep learning
Traditional financial risk early warning has problems such as data processing difficulties and single data.In order to solve these problems,the study proposes a multilayer feed-forward neural network(Back Propagation,BP)financial risk early warning system model based on the symbolic rank sum test.The model uses BP neural network to predict financial data and symbolic rank sum test to identify the type and size of financial risks.The model algorithm was compared with Recurrent Neural Network(RNN)and Convolutional Neural Networks(CNN).The results show that the area under the line of the Receiver Operating Characteristic Curve(ROC)curve of the model algorithm is 0.86,the mean absolute error average and the detection rate average are 2.67%and 86.13%,respectively,which are the optimal values,and the model algorithm's early warning capacity in the training set is 80.7%.It is confirmed that the model algorithm proposed in the study can improve the accuracy and reliability of financial risk early warning,effectively predict financial risks,and provide an important reference for financial risk management and decision-making.
万方数据
维普
