Forecasting Indonesia's Economic Development During the U.S.–China Trade War: A Machine Learning Approach Using Macroeconomic Determinants

This study applies an ensemble of supervised machine learning (ML) models—Linear Regression, Decision Tree, Random Forest, and Gradient Boosting—to forecast Indonesia’s GDP growth using 120 monthly macroeconomic observations spanning January 2016 to December 2025. The dataset incorporates trade balance, unemployment rate, inflation, government revenue, net imports, net exports, and the Jakarta Composite Index (IHSG) year-on-year returns, supplemented by temporal dummy variables to capture structural breaks and seasonal effects. A second model specification additionally incorporates proxies for Indonesia’s hilirisasi downstream industrial policy and a bilateral trade tension index.