Optimization of Indonesia's Trade Predictions Through Penalized Regression Models

Economic activities are important in a country's economic movements. Some economic activities that influence this movement include export and import activities. Therefore, this study analyzes Indonesia's export and import activities using macroeconomic factors such as Gross Domestic Product (GDP), foreign exchange rates, money supply, inflation rate, Wholesale Price Index (WPI), gold futures prices, and stock indices. Due to the presence of multicollinearity in the data, the analysis is conducted using Regularized Least Squares Regression (RIDGE), Least Absolute Shrinkage and Selection Operator Regression (LASSO), and Elastic Net regression which are penalized regression techniques that can enhance and optimize the prediction for Indonesia's export and import values. Furthermore, the performance of these three models is evaluated using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The results show that the LASSO method provided the best result for import, oil and gas export, and non-oil and gas export values. With a testing MAPE of 12.0224% and a testing RMSE of 2411.3428 for imports. A testing MAPE of 59.0001% and a testing RMSE of 846.6390 for oil and gas exports. Lastly, a testing MAPE of 9.2385% and a testing RMSE of 2321.6906 for non-oil and gas exports.