A study on asset pricing in stock market based on Hopfield neural network

Nowadays, with globalisation and financial globalisation, stock markets are becoming more and more complex, which cannot be explained by classical financial analysis. In this paper, Hopfield neural network is used to describe the complex nonlinear and asymmetric financial system in detail, and a stock market prediction method is proposed. The specific results are as follows: Hopfield neural network is introduced, and the network structure, operation mode and convergence principle are described in detail. Finally, the performance of FNN is successfully simulated. On the basis of the simulation model, we used the Hopfield neural network based on FSK and applied it to the FSK based on RFE-GSWOA-Hopfield; the MAPE value, the RMSE value, and the MAE value of the model proposed in this paper reached the minimum values in several models, which were 29.206, 0.594,23.131, and the R2 reaches the maximum value of 0.952 in several models.In addition, the experiments also prove that the RFE-GSWOA-Hopfield model is better in prediction efficiency and optimisation accuracy.