Perbandingan Metode Arima (Box Jenkins), Multiscale Autoregressive (MAR), dan Singular Spectrum Analysis (SSA) untuk Data Non-Stationer dalam Peramalan Data Nilai Ekspor Provinsi Bengkulu FOB (Free On Board) Pelabuhan Baai Januari 2019 - September 2023

Shidigie, A A
Yurike, L
Puspita, D
Julieta, A
Hidayati, N
Putri, M H C

Diophantine Journal of Mathematics and Its Applications
Vol. 3 No. 1 (2024)

This research presents a comparison of the performance of three forecasting methods, namely ARIMA (Box Jenkins), Multiscale Autoregressive (MAR), and Singular Spectrum Analysis (SSA), in dealing with non-stationary export data challenges. The focus of the study is on forecasting the export value of Bengkulu Province FOB (Free on Board) Pelabuhan Baai from January 2019 to September 2023. By using ARIMA as a classical approach, MAR and SSA as representations of multiscale and signal decomposition approaches, this study aims to provide a comprehensive understanding of the effectiveness of each method in dealing with dynamic export data characteristics. Performance evaluation is carried out using criteria such as Mean Absolute Percentage Error (MAPE), with the hope of providing valuable insights for selecting the optimal forecasting method in the context of Bengkulu Province's exports.