EUR-USD Exchange Rate Forecasting Based on Information Fusion with Large Language Models and Deep Learning Methods

The exchange rate between the Euro and the US Dollar is a significant indicator in the global financial market, reflecting the economic dynamics between two of the world's largest economies. Precise prediction of the EUR/USD exchange rate is crucial for individual investors, businesses engaged in international trade, and policymakers responsible for economic stability and growth. Traditionally, econometric models have been utilized to forecast exchange rates, relying heavily on historical market data and macroeconomic indicators released by governments and financial organizations [1]. Although these datasets are comprehensive, their low publication frequency makes it difficult to capture real-time market volatility and nonlinear dynamics [2]. The integration of unstructured data from diverse sources, such as news articles, financial reports and social media platforms, has the potential to improve the accuracy of exchange rate forecasting. In recent years, the significant impact of political events, global economic crises, and unexpected international incidents on currency fluctuations has been recognized [3], suggesting that considering a wider range of information beyond traditional structured data may be beneficial. The great amount of textual data may contain valuable insights into market sentiment, economic trends, and key events that can influence exchange rates [4]. However, exchange rate forecasting presents two significant challenges. Firstly, while the relationship between news information and market trends is relatively straightforward in traditional financial markets, the complex semantics of market-driven news and analysis texts in the context of exchange rates pose difficulties for sentiment analysis.