Komparasi Metode Regresi Linier, Exponential Smoothing dan ARIMA Pada Peramalan Volume Ekspor Minyak Kelapa Sawit di Indonesia

Abstract

This study aims to compare several methods to get the best methods on forecasting the volume of Indonesian palm oil exports. In addition, this study also aims to estimate the volume of Indonesian palm oil exports for the next five years. Some of the forecasting methods used in this study are linear regression, exponential smoothing, and ARIMA. The data used is historical data on the volume of palm oil exports from 1981 to 2020. The results of calculations and analysis show that the exponential smoothing model of the damped trend method produces the smallest error value compared to other methods, the MAD value is 860,353, the MSE value is 1,707,738,707,222, the RSME value is 1,306,805, and the MAPE value is 20.6%. This method has chosen to be the best forecasting method for the next five years. The forecast results obtained that the volume of Indonesian palm oil exports for the next five years are28.864.223,31 tons, 28.967.062,92 tons, 29.064.976,80 tons, 29.158.200,89 tons, and 29.246.959,81 tons.