Forecasting the Change of the US Dollar Price against the Chinese Yuan Based on Neural Network Modeling

Under the background of intensified economic globalization, the exchange rate of the US dollar against the RMB is a major issue in the current economic and trade relations between China and the United States and even in the world's financial field. Consequently, accurately forecasting the movement of the US dollar in relation to the Chinese yuan holds significant practical value. Earlier research primarily employed various quantitative modeling techniques to analyze the time-series characteristics of the USD/CNY exchange rate. However, inconsistencies remain regarding specific issues, necessitating the selection of different models tailored to each problem. As the BP neural network has the advantages of self-learning, self-adaptation, anti-interference, and high generalization. By optimizing the parameters of neural network and improving the data set, an optimization method of the number of hidden layer neurons based on Bayesian algorithm is proposed in this paper. Through the training results and analysis of the neural network, it can be seen that using the improved BP neural network to forecast the exchange rate of RMB against US dollar has higher accuracy and effectiveness and can predict the exchange rate changes of US dollar against RMB more accurately. In real life, it has great reference value for investors to formulate investment strategy and enterprises to carry out risk management.