Artificial intelligence for macroeconomic stability: An interpretable machine learning approach for the real exchange rate in Bolivia
Fecha de Publicación
Autores
Julio Cesar Nava León
Número Especial de Machine Learning - Cuadernos de Investigación Económica Boliviana (2024) Vol. 7(2), 24-42
Resumen
This study employs advanced machine learning techniques—Random Forest, XGBoost, LightGBM, and CatBoost—to estimate the equilibrium real exchange rate (ERER) in Bolivia from January 1992 to May 2024. Using a comprehensive dataset of 445 macroeconomic and climatic variables, including production, financial indicators, domestic prices, and com modity prices, the analysis captures complex nonlinear dynamics in exchange rate behavior.