BP neural network predicts the exchange rate of US dollar against gold

Abstract

Gold and the US dollar are important currencies, and their price changes reflect changes in the world and North American economies. The gold market has become a financial investment market of equal importance to markets such as stocks, foreign exchange, and bonds. In such a situation, the price change of gold has become an important predictive research object. However, due to the changing world landscape, the randomness of gold price changes, and the existence of exchange rate and other influencing factors, it is very difficult to predict. It is crucial to establish a prediction model for exchange rate changes over time. This topic establishes a neural network model for exchange rate changes over time using the newly developed Back Propagation Neural Network in recent years. Finally, after several experiments, the parameters are adjusted to the optimal state to obtain the minimum error result. Through this study, we can determine that the prediction function of the BP artificial neural network can be applied in certain fields of life, but it also has significant limitations, to make it a complete tool for our daily life or work, it is necessary to further improve this technology.