Exchange Rate Prediction: Micro and Macro Factors, Machine Learning, and Future Directions

Abstract

This review synthesizes current research on the dynamics of exchange rates, examining the influence of microeconomic factors, such as market sentiment, and macroeconomic factors, including political stability and economic governance. It also explores the application of modern predictive techniques, notably machine learning and deep learning technologies, in anticipating exchange rate movements. The review emphasizes how these cutting-edge methods can be used to better foresee and capture intricate nonlinear interactions, while also underscoring the importance of considering both microeconomic and macroeconomic factors in developing effective prediction models. In the end, this analysis highlights significant patterns and gaps in the literature, offering insightful information to investors, politicians, and scholars attempting to understand the nuances of exchange rates. It concludes by emphasizing the need for future research to address existing limitations in data acquisition and model selection, as well as exploring new avenues for improving exchange rate prediction, such as incorporating diverse data sources and developing more robust and adaptive models.