PEMODELAN AUTOREGRESSIVE DISTRIBUTED LAG UNTUK MEMPREDIKSI NILAI IMPOR NON-MIGAS DI INDONESIA

Abstract

The International Monetary Fund warns countries about the global economic recession in 2023. Efforts required from policy makers to prevent a recession. A deficit balance of payments shows signs of recession because the rate of imports is higher than exports. The highest import value over the last decade is non-oil and gas commodities. Factors affecting imports include exchange rates, prices of goods, and consumer income. Import activities require proper studies to make policies so that research is needed, one of which is by using the Autoregressive Distributed Lag (ARDL) method. ARDL is a regression model in which the independent variable consists of the current and past independent variable values and the past values of the dependent variable. The data used is from the first quarter of 2008 to the fourth quarter of 2022. The model formed is ARDL(3,2,2,1). The current non-oil and gas import is positively affected by the non-oil and gas imports at lag one, three, and four, as well as the exchange rate at lag three and is negatively affected by the non-oil and gas import and the exchange rate at lag two. The accuracy of forecasting with sMAPE is 12,12%, which means the forecast is accurate.