COMPARISON OF SEVERAL DOUBLE EXPONENTIAL SMOOTHING METHODS ON FORECASTING RICE PRICE IN AMBON CITY

Abstract

Rice is one of the staple foods consumed daily by the Indonesian people. Maluku Province's rice production ranks 28^th out of 38 provinces in Indonesia in 2024. However, as the provincial capital, Ambon City does not have any rice production at all, resulting in a high price difference between rice and other regions. This study aims to compare the accuracy of the Double Exponential Smoothing Brown Method, Double Exponential Smoothing Holt and Damped Trend and to obtain the results of rice price forecasting in Ambon City using the best model. The methods used are Brown's Double Exponential Smoothing method, Holt's Double Exponential Smoothing method, and the Damped Trend method. The results of this study indicate that the Damped Trend method with parameters α = 0.9, β = 0.9, and φ = 0.4 is the best model with a MAPE value of 0.759%. The forecast results for the period September to December 2025 are respectively IDR 17,796/kg, IDR 17,839/kg, IDR 17,857/kg, and IDR 17,864/kg. This shows that the price of rice in Ambon City in the period from September to December 2025 will experience a relatively small increase.