<![CDATA[Research Originality: While numerous studies have examined exchange rate determinants in Indonesia, limited evidence integrates key macroeconomic variables within a unified framework while comparing conventional econometric and machine learning approaches. Research Objectives: This study examines the effects of inflation, interest rates, money supply, trade balance, and foreign exchange reserves on the Indonesian Rupiah exchange rate and compares the forecasting performance of VECM and machine learning models. Research Methods: A Vector Error Correction Model (VECM) is employed to analyze long-run and short-run relationships, while Support Vector Regression (SVR), Random Forest, and Long Short-Term Memory (LSTM) are used for forecasting. Monthly data from January 2010 to December 2024 are analyzed. Forecasting performance is evaluated using Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE). Empirical Results: The results indicate a long-run cointegrating relationship between exchange rates and macroeconomic fundamentals. In the short run, money supply significantly affects exchange rate movements. Among the forecasting models, LSTM achieves the highest predictive accuracy based on MAE, MAPE, and RMSE. Implications: The findings highlight the importance of macroeconomic fundamentals in maintaining exchange rate stability and demonstrate the potential of machine learning techniques, particularly LSTM, for exchange rate forecasting in Indonesia. JEL Classification: C32, C45, E44, F31 How to Cite:Akbar, A. F. (2026). What Drives the Indonesian Rupiah? Evidence from VECM and Machine Learning. Signifikan: Jurnal Ilmu Ekonomi, 15(2), 571-584. https://doi.org/10.15408/sjie.v15i2.51329.]]>
