Analysis and Forecast of Exchange Rate Based on Reinforcement Learning

Abstract

For individuals, exchange rate changes affect individuals' consumption of foreign goods and services; for government enterprises, exchange rate forecasting is the basis for corporate foreign exchange risk measurement and strategic adjustment. Therefore, exchange rate forecast analysis has importance research value. Based on the specific situation of exchange rate trading under big data, this paper analyzes and predicts the RMB exchange rate. At the same time, by using multiple reinforcement learning models, it is predicted when and what kind of buy-and-sell strategy can finally achieve effective returns. Finally, the model results of the model are compared between multiple currency exchanges, and the model is compared and demonstrated. The validity and adaptability ensure the accuracy of the model. The results of this paper model can be used as a reference for individuals and even the government for exchange rate changes.