Predicting USD to IDR Exchange Rates with Decision Trees

Abstract

Predicting currency exchange rates is a complex challenge due to the numerous factors influencing market fluctuations. This study explores the application of decision trees to predict the USD to IDR exchange rate, leveraging historical data and key economic indicators. Decision trees, known for their ability to model non-linear relationships, offer an interpretable approach to understanding the factors driving exchange rate movements. The study demonstrates that decision trees can successfully capture the patterns in the data, providing a foundation for accurate predictions. However, the volatility and unpredictability of exchange rates, driven by geopolitical events, market sentiment, and macroeconomic shifts, highlight the limitations of the model. While decision trees provide a valuable starting point, the research suggests that combining them with advanced methods, such as ensemble techniques (random forests or gradient boosting) or time-series models (ARIMA or LSTM), could improve forecasting accuracy. Incorporating a wider range of features, including macroeconomic indicators and market sentiment analysis, further enhances the model's robustness. The findings underscore the need for hybrid approaches that combine the strengths of multiple models to better capture the dynamic and complex nature of financial markets. This research contributes to the broader understanding of exchange rate prediction and offers practical insights for businesses and financial institutions seeking to make informed decisions.