Systematic framework for ARIMA and SARIMA model selection using multiple error metrics: An application to export forecasting
Article Sidebar
Main Article Content
Abstract
This study proposes a systematic and practical framework for selecting autoregressive integrated moving average (ARIMA) and seasonal autoregressive integrated moving average (SARIMA) forecasting models based on multiple error evaluation metrics, including root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE). In practical forecasting applications, different error metrics often produce conflicting model selection outcomes, resulting in uncertainty in decision-making. Accordingly, this study introduces a percentage-based consistency evaluation approach to identify the most reliable forecasting model across multiple datasets. This study proposes a systematic framework for ARIMA and SARIMA model selection that integrates multiple forecasting error metrics to address conflicting model rankings and improve the consistency of model-selection decisions. This framework is applied to Thailand’s export value data, consisting of 69 monthly observations across 12 datasets derived from major industrial product groups (HS-84, HS-85, and HS-87) and key trading partners, including China, Japan, and South Korea. Results indicate that ARIMA(1,2,1) achieves the highest level of consistency across error metrics, with 83.33% under RMSE and 75% under MAPE and MAE. In terms of seasonal models, SARIMA (2,2,2)(2,0,0)₁₂ demonstrates the best overall performance. The findings confirm that the choice of error metrics can significantly influence forecasting outcomes. The proposed consistency-based framework reduces model selection conflicts and enhances the reliability of decision-making. This approach provides practical guidance for forecasting applications in economic planning, industrial systems, and export management.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
Heizer J, Render B, Munson C. Operations management: sustainability and supply chain management. 13th ed. Harlow: Pearson; 2020.
Makridakis S, Spiliotis E, Assimakopoulos V. The M4 Competition: 100,000 time series and 61 forecasting methods. Int J Forecast. 2020;36(1):54-74. DOI: https://doi.org/10.1016/j.ijforecast.2019.04.014
Hyndman RJ, Athanasopoulos G. Forecasting: principles and practice. 3rd ed. Melbourne: OTexts; 2021.
Masini RP, Medeiros MC, Mendes EF. Machine learning advances for time series forecasting. J Econ Surv. 2023;37(1):76-111. DOI: https://doi.org/10.1111/joes.12429
Hutasuhut AH, Anggraeni W, Tyasnurita R. Development of a decision support application for forecasting raw material inventory using the ARIMA (autoregressive integrated moving average) method at CV Asia. Jurnal Teknik Pomits. 2014;3(2):169-74. (In Indonesian)
Silalahi DK. Forecasting of poverty data using seasonal ARIMA modeling in West Java Province. Jurnal Teori dan Aplikasi Matematika. 2020;4(1):76-86. DOI: https://doi.org/10.31764/jtam.v4i1.1888
Sankalp S, Panda PK. A comparative evaluation of machine learning and ARIMA models for forecasting relative humidity over Odisha districts. In: Kasiviswanathan KS, Soundharajan B, Patidar S, He J, Ojha CSP, editors. Modeling and Mitigation Measures for Managing Extreme Hydrometeorological Events Under a Warming Climate. Developments in Environmental Science. Vol. 14. Amsterdam: Elsevier; 2023. p. 91-105. DOI: https://doi.org/10.1016/B978-0-443-18640-0.00013-4
Mgaya JF. Application of ARIMA models in forecasting livestock products consumption in Tanzania. Cogent Food Agric. 2019;5(1):1607430. DOI: https://doi.org/10.1080/23311932.2019.1607430
Chyon FA, Suman MNH, Fahim MRI, Ahmmed MS. Time series analysis and predicting COVID-19 affected patients by ARIMA model using machine learning. J Virol Methods. 2022;301:114433. DOI: https://doi.org/10.1016/j.jviromet.2021.114433
Eiglsperger J, Haselbeck F, Grimm DG. ForeTiS: A comprehensive time series forecasting framework in Python. Mach Learn Appl. 2023;12:100467. DOI: https://doi.org/10.1016/j.mlwa.2023.100467
Ariyanti VP, Yusnitasari T. Comparison of ARIMA and SARIMA for forecasting crude oil prices. Jurnal Resti. 2023;7(2):405-13. DOI: https://doi.org/10.29207/resti.v7i2.4895
Ning Y, Kazemi H, Tahmasebi P. A comparative machine learning study for time series oil production forecasting: ARIMA, LSTM, and Prophet. Comput Geosci. 2022;164:105126. DOI: https://doi.org/10.1016/j.cageo.2022.105126
Suppalakpanya K, Nikhom R, Booranawong T, Booranawong A. Forecasting oil palm and crude palm oil data in Thailand using exponential time-series methods. Eng Appl Sci Res. 2019;46(1):44-55.
Venkataraman V, Prasad S, Aswathanarayana B, Barigidad S, Nayak V, Kumar NST. Development of time series models for various pollutants in Bangalore city using the Akaike information criterion. Eng Appl Sci Res. 2020;47(3):249-63.
DeLurgio SA. Forecasting principles and applications. New York: McGraw-Hill; 1998.
Demir L, Akkaş S. A comparison of sales forecasting methods for a feed company: a case study. Pamukkale University Journal of Engineering Sciences. 2018;24(4):705-12. DOI: https://doi.org/10.5505/pajes.2018.58235
Zhang Y, Na S. A novel agricultural commodity price forecasting model based on fuzzy information granulation and MEA-SVM model. Math Probl Eng. 2018;2018(1):1-10. DOI: https://doi.org/10.1155/2018/2540681
Athipayoon N, Thongteeraparp A, Chaipitak S. Comparison of forecasting methods for non-linear and non-stationary time series. Thai Science and Technology Journal (TSTJ). 2020;28(2):197-207. (In Thai)
Poler R, Mula J. Forecasting model selection through out-of-sample rolling horizon weighted errors. Expert Syst Appl. 2011;38(12):14778-85. DOI: https://doi.org/10.1016/j.eswa.2011.05.072
Badulescu Y, Hameri AP, Cheikhrouhou N. Evaluating demand forecasting models using multi-criteria decision-making approach. J Adv Manag Res. 2021;18(5):661-83. DOI: https://doi.org/10.1108/JAMR-05-2020-0080
Koubaa Z, El-Amraoui A, Frikha A, Delmotte F. Multicriteria decision making for selecting forecasting electricity demand models. Sustainability. 2024;16(21):9219. DOI: https://doi.org/10.3390/su16219219
Stroilă AM. Enhancing univariate time series forecasting: a general tree-based model selection approach. Comput Stat. 2026;41:46. DOI: https://doi.org/10.1007/s00180-025-01703-5
Mahanil N, Khemavuk P. A conceptual model for developing a selection error measurement method to measure the effectiveness of the time series forecasting method. 13th Annual International Conference on Industrial Engineering and Operations Management; 2023 Mar 7-9; Manila, Philippines. Manila: IEOM; 2023. p. 1992-2003. DOI: https://doi.org/10.46254/AN13.20230545
European Commission. Harmonised system [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://taxation-customs.ec.europa.eu/customs/customs-tariff/tariff-classification-goods/harmonised-system_en.
Trading Economics. Thailand exports [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://tradingeconomics.com/thailand/exports.
Tendata. Thailand export data and top export products [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://tradereport.moc.go.th/en/stat/report02?utm_source=chatgpt.com
Box GEP, Jenkins GM, Reinsel GC, Ljung GM. Time series analysis: forecasting and control. 5th ed. Hoboken: Wiley; 2015.
Hyndman RJ, Khandakar Y. Automatic time series forecasting: the forecast package for R. J Stat Softw. 2008;27(3):1-22. DOI: https://doi.org/10.18637/jss.v027.i03
Hyndman RJ, Athanasopoulos G. Forecasting: principles and practice. 2nd ed. Melbourne: OTexts; 2018. DOI: https://doi.org/10.32614/CRAN.package.fpp2
McKinney W. Data structures for statistical computing in Python. Proceedings of the 9th Python in Science Conference (SciPy 2010); 2010 Jun 28 – Jul 3; Austin, Texas. p. 56-61. DOI: https://doi.org/10.25080/Majora-92bf1922-00a
Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, et al. Array programming with NumPy. Nature. 2020;585:357-62. DOI: https://doi.org/10.1038/s41586-020-2649-2
Seabold S, Perktold J. Statsmodels: econometric and statistical modeling with Python. Proceedings of the 9th Python in Science Conference (SciPy 2010); 2010 Jun 28 – Jul 3; Austin, Texas. p. 92-6. DOI: https://doi.org/10.25080/Majora-92bf1922-011
James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning: with applications in R. 2nd ed. New York: Springer; 2021. DOI: https://doi.org/10.32614/CRAN.package.ISLR2
Dickey DA, Fuller WA. Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc. 1979;74(366):427-31. DOI: https://doi.org/10.1080/01621459.1979.10482531
National Institute of Standards and Technology (NIST). Box-Jenkins model identification [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://www.itl.nist.gov/div898/handbook/pmc/section4/pmc446.htm.
Hamilton JD. Time series analysis. Princeton: Princeton University Press; 1994.
Brockwell PJ, Davis RA. Introduction to time series and forecasting. 3rd ed. Cham: Springer; 2016. DOI: https://doi.org/10.1007/978-3-319-29854-2
Taylor BW, Russell RS. Operations management: creating value along the supply chain. 7th ed. Hoboken: John Wiley & Sons; 2011.
Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE). Clim Res. 2005;30(1):79-82. DOI: https://doi.org/10.3354/cr030079
de Myttenaere A, Golden B, Le Grand B, Rossi F. Mean absolute percentage error for regression models. Neurocomputing. 2016;192:38-48. DOI: https://doi.org/10.1016/j.neucom.2015.12.114
Shalabi LA, Shaaban Z, Kasasbeh B. Data mining: a preprocessing engine. J Comput Sci. 2006;2(9):735-9. DOI: https://doi.org/10.3844/jcssp.2006.735.739
Han J, Kamber M, Pei J. Data mining: concepts and techniques. 3rd ed. Waltham: Morgan Kaufmann; 2012.
Yoon K. A reconciliation among discrete compromise solutions. J Oper Res Soc. 1987;38(3):277-86. DOI: https://doi.org/10.1057/jors.1987.44
Çelen A. Comparative analysis of normalization procedures in TOPSIS method: With an application to Turkish deposit banking market. Informatica. 2014;25(2):185-208. DOI: https://doi.org/10.15388/Informatica.2014.10
Thorndike RL. Who belongs in the family?. Psychometrika. 1953;18:267-76. DOI: https://doi.org/10.1007/BF02289263
MacQueen J. Some methods for classification and analysis of multivariate observations. Berkeley Symp Math Statist Prob. 1967:281-97.
Jain AK. Data clustering: 50 years beyond K-means. Pattern Recognit Lett. 2010;31(8):651-66. DOI: https://doi.org/10.1016/j.patrec.2009.09.011
Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction. 2nd ed. New York: Springer; 2009. DOI: https://doi.org/10.1007/978-0-387-84858-7
General Administration of Customs of the People’s Republic of China (GACC). China customs statistics [Internet]. 2026 [cited 2026 Apr 4]. Available from: http://english.customs.gov.cn/.
Japan External Trade Organization (JETRO). Japan trade statistics [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://www.jetro.go.jp/en/reports/statistics.html.
Korea Customs Service (KCS). Trade statistics [Internet]. 2026 [cited 2026 Apr 4]. Available from: https://unipass.customs.go.kr.
Wu D. A comparative analysis of ARIMA models for forecasting China’s GDP. Adv Oper Res Prod Manag. 2025;4(2):1-11. DOI: https://doi.org/10.54254/3029-0880/2025.26616
Dlamini WJ, Melesse SF, Mwambi HG. A comparative study of ARIMA, ARFIMA, and ARIMA-ARFIMA models in predicting under-five mortality rate in four East-African Nations. Open Public Health J. 2026;19:1-10. DOI: https://doi.org/10.2174/0118749445420131260114041333
Yang W, Su A, Ding L. Application of exponential smoothing method and SARIMA model in predicting the number of admissions in a third-class hospital in Zhejiang Province. BMC Public Health. 2023;23(1):2309. DOI: https://doi.org/10.1186/s12889-023-17218-x
