Perbandingan Kinerja Model Forecasting Nilai Perdagangan Komoditas HS pada Evaluasi Time-Based
DOI:
https://doi.org/10.30865/jurikom.v13i3.9843Keywords:
- Forecasting, Time-based, Industry, SARIMA, HS CommoditiesAbstract
Global economic uncertainty, trade-regime shifts, and supply-chain disruptions have made export-import trade-value forecasting increasingly complex. This study compares the performance of Random Forest, Extra Trees Regression, and SARIMA in predicting monthly trade values of HS commodities in the apparel and footwear sector, covering HS 61, HS 62, HS 63, and HS 64. The dataset was obtained from Indonesia?s Central Bureau of Statistics (BPS) and organized as a monthly time series using a leakage-safe workflow through a time-based train-validation-test split. The modeling stage employed 19 predictive features consisting of historical, local statistical, calendar-seasonal, and exogenous variables, and each model was tuned on the validation set before being evaluated on the holdout test. Performance was assessed using MAE, RMSE, MAPE, sMAPE, and wMAPE, with MAPE as the primary ranking metric. The main contribution of this study lies in providing a fair and replicable comparison of three forecasting models under a time-based evaluation protocol for HS commodity trade data, making the model selection results more representative of real implementation settings. The results show that Random Forest achieved the best MAPE at 22.7746%, slightly outperforming Extra Trees Regression at 22.9469%, while SARIMA recorded 28.9794%. These findings indicate that tree-based ensemble models are more adaptive to volatile trade data, whereas SARIMA remains relevant as a statistical baseline for structured seasonal patterns.
References
[1] D. Rincon-Yanez, C. Ounoughi, B. Sellami, T. Kalvet, M. Tiits, S. Senatore, and S. Ben Yahia, “Accurate prediction of international trade flows: Leveraging knowledge graphs and their embeddings,” J. King Saud Univ. Comput. Inf. Sci., vol. 35, p. 101789, 2023, doi: 10.1016/j.jksuci.2023.101789.
[2] C. Arndt, X. Diao, P. Dorosh, K. Pauw, and J. Thurlow, “The Ukraine war and rising commodity prices: Implications for developing countries,” Glob. Food Secur., vol. 36, p. 100680, 2023, doi: 10.1016/j.gfs.2023.100680.
[3] F. Lin et al., “The impact of Russia-Ukraine conflict on global food security,” Glob. Food Secur., vol. 36, p. 100661, 2023, doi: 10.1016/j.gfs.2022.100661.
[4] X. Song et al., “Deep learning-based time series forecasting: A comprehensive review,” Artif. Intell. Rev., 2024, doi: 10.1007/s10462-024-10989-8.
[5] B. Sellami et al., “Harnessing Graph Neural Networks to predict international trade flows,” Data, vol. 9, no. 6, 2024, doi: 10.3390/data9060065.
[6] I. Chowdhury, Z. Kun, and G. Boako, “Geopolitical risk and energy market tail risk forecasting,” J. Commod. Mark., vol. 25, p. 100334, 2025, doi: 10.1016/j.jcomm.2025.100334.
[7] P. Wang et al., “Interval time series forecasting: A systematic literature review,” J. Forecast., 2024, doi: 10.1002/for.3024.
[8] T. Hothorn et al., “Probabilistic time series forecasts with autoregressive transformation models,” Stat. Comput., 2023, doi: 10.1007/s11222-023-10212-8.
[9] F. Schlembach, E. Smirnov, I. Koprinska, et al., “Conformal multistep-ahead multivariate time-series forecasting,” Mach. Learn., vol. 114, p. 165, 2025, doi: 10.1007/s10994-024-06722-9.
[10] J. Schosser et al., “Beyond point forecasts: Uncertainty quantification in tensor extrapolation for relational time series data,” J. Bus. Anal., 2024, doi: 10.1080/23737484.2024.2329171.
[11] S. Allen, J. Ziegel, and D. Ginsbourger, “Assessing the calibration of multivariate probabilistic forecasts,” Q. J. R. Meteorol. Soc., 2024, doi: 10.1002/qj.4647.
[12] M. Sousa, A. M. Tome, and J. Moreira, “A general framework for multi-step ahead adaptive conformal heteroscedastic time series forecasting,” Neurocomputing, vol. 608, p. 128434, 2024, doi: 10.1016/j.neucom.2024.128434.
[13] S. Allen, J. F. Ziegel, and T. Gneiting, “Tail calibration of probabilistic forecasts,” J. Am. Stat. Assoc., 2025, doi: 10.1080/01621459.2025.2506194.
[14] Z. Senousy, M. M. Gaber, and M. M. Abdelsamea, “AUQantO: Actionable uncertainty quantification optimization in deep learning architectures,” Appl. Soft Comput., vol. 146, p. 110666, 2023, doi: 10.1016/j.asoc.2023.110666.
[15] H. Turbe et al., “Evaluation and explanation of deep learning-based time series forecasting models,” Nat. Mach. Intell., 2023, doi: 10.1038/s42256-023-00724-9.
[16] M. Shajalal, A. Boden, and G. Stevens, “ForecastExplainer: Explainable household energy demand forecasting by approximating Shapley values using DeepLIFT,” Technol. Forecast. Soc. Change, vol. 206, p. 123588, 2024, doi: 10.1016/j.techfore.2024.123588.
[17] W. Jo and H. Kim, “Neural additive time-series models: Explainable deep learning for time series forecasting,” Expert Syst. Appl., vol. 223, p. 119954, 2023, doi: 10.1016/j.eswa.2023.119954.
[18] A. Pesenti and A. O’Sullivan, “Explaining deep neural network models for electricity price forecasting with explainable AI,” Energy AI, vol. 21, p. 100532, 2025, doi: 10.1016/j.egyai.2025.100532.
[19] A. Neubauer, S. Brandt, and M. Kriegel, “Explainable multi-step heating load forecasting: Using SHAP values and temporal attention mechanisms for enhanced interpretability,” Energy AI, vol. 20, p. 100480, 2025, doi: 10.1016/j.egyai.2025.100480.
[20] N. Wang, W. Shao, et al., “Carbon trading price forecasting with a new explainable machine learning approach,” Technol. Forecast. Soc. Change, vol. 200, p. 123178, 2024, doi: 10.1016/j.techfore.2024.123178.
[21] X. Wu and J. Wang, “Wind speed forecasting using Temporal Fusion Transformers and SHAP,” Energy Rep., vol. 10, pp. 5213–5224, 2024, doi: 10.1016/j.egyr.2023.12.073.
[22] J. Xia et al., “A probability forecasting model for cotton yarn futures price based on LassoNet and explainable AI,” J. Forecast., 2025, doi: 10.1002/for.3185.
[23] Y. Xu, Z. Liu, et al., “Crude oil futures volatility forecasting using a Bi-LSTM-based deep learning approach,” Resour. Policy, vol. 80, p. 103098, 2024, doi: 10.1016/j.resourpol.2023.103098.
[24] O. Celik, E. Altinsoy, A. Suliman, et al., “Hybrid forecasting for agricultural commodity prices: Statistical learning and interpretable machine learning evidence,” Borsa Istanbul Rev., 2025, doi: 10.1016/j.bir.2025.04.003.
[25] A. Bara et al., “Interpretable transformer models for predicting electricity market prices based on interpretable encoding,” Expert Syst. Appl., p. 126118, 2025, doi: 10.1016/j.eswa.2025.126118.
[26] L. M. Dang et al., “Toward explainable heat load patterns prediction for district heating networks,” Sci. Rep., 2023, doi: 10.1038/s41598-023-34146-3.
[27] M. Ardiyanto, R. D. Irawan, and K. A. Yudhianto, “Implementation and optimization of saliency mapping algorithms in convolutional neural networks to enhance transparency in pneumonia diagnosis,” in Proc. Int. Conf. Sci. Health Technol., 2025, pp. 490–500.
[28] S. Levich et al., “Incorporating digital documents into predictive process monitoring using deep neural networks,” Decis. Support Syst., vol. 175, p. 114043, 2023, doi: 10.1016/j.dss.2023.114043.
[29] R. D. Irawan, M. Adha, M. P. Sadana, Z. D. K. W. Arba’ah, and E. Utami, “Pemodelan hasil rekayasa kebutuhan perangkat lunak sistem jurnal elektronik terintegrasi IDEOGRAM,” JIKO J. Inform. dan Komput., vol. 7, no. 1, p. 13, 2023.
[30] R. D. Irawan and A. Fatkhurohman, “Application of deep learning algorithm to detect fraud in online transaction networks,” J. Teknol. dan Open Source, vol. 7, no. 2, pp. 167–177, 2024.
[31] P. Pramono, R. D. Irawan, and A. Sari, “Comparative analysis of SQL injection attack classification using Naive Bayes method and Support Vector Machine,” in Proc. Int. Conf. Sci. Health Technol., 2024, pp. 208–217.
[32] H. Chen et al., “Explainable machine learning methods for predicting building energy consumption,” Sci. Rep., 2025, doi: 10.1038/s41598-025-27285-2.
[33] M. Abdullah, “Explainable deep learning model for stock price forecasting using textual analysis,” Expert Syst. Appl., 2024, doi: 10.1016/j.eswa.2024.123740.
[34] S. Gunasekaran et al., “A predictive approach to enhance time-series forecasting,” Nat. Commun., 2025, doi: 10.1038/s41467-025-63786-4.
[35] X. Ge et al., “STGATN: A wind speed forecasting method based on spatial dependency,” Int. J. Digit. Earth, 2025, doi: 10.1080/17538947.2025.2496794.
[36] R. J. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice, 3rd ed. Melbourne, Australia: OTexts, 2021.
[37] C. Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. 2025.
