Studi Komparatif Model Double Exponential Smoothing Brown dan ARIMA Dalam Mengevaluasi Proyeksi Kinerja Ekspor Indonesia Periode 2020-2025

Renata Cecilia Br Sirait
Ahda Azkiya Susilo
Faizal Ahmad Habibie
Muh. Aryo Kusuma Hidayat
Neli Agustina

Prosiding Seminar Nasional Official Statistics
Vol 2026 No 1 (2026): Seminar Nasional Official Statistics 2026

Oil and non-oil exports are key indicators of Indonesia's international trade performance and experienced fluctuations during the 2020–2025 period. This study compares the forecasting accuracy of the Double Exponential Smoothing Brown (DES Brown) and Autoregressive Integrated Moving Average (ARIMA) models in projecting Indonesia's export performance. Monthly export volume data from Statistics Indonesia (BPS) covering January 2020 to December 2025 were used. The research involved parameter optimization for the DES Brown model and ARIMA model identification. The results show that DES Brown with a smoothing parameter of α = 0.1 achieved better forecasting accuracy with a MAPE of 8.51%, while the best ARIMA model, ARIMA(1,1,1), produced a MAPE of 9.868%. The forecast indicates an upward trend in Indonesia's export volume, reaching 67,910.39 thousand tons in December 2026.