Comparison of Linear Regression Method and Holt-Winters Method in Oil and Gas Export Volume Forecasting in Indonesia

Nuraini Sibuea; Andi Rio Ramadhan; Ratna Tri Aulia; Siska Khairunnisa; Rado Yendra; Arisman Adnan

doi:https://doi.org/10.14445/22315373/IJMTT-V68I12P502

International Journal of Mathematics Trends and Technology

Research Article | Open Access | Download PDF

Volume 68 | Issue 12 | Year 2022 | Article Id. IJMTT-V68I12P502 | DOI : https://doi.org/10.14445/22315373/IJMTT-V68I12P502

Comparison of Linear Regression Method and Holt-Winters Method in Oil and Gas Export Volume Forecasting in Indonesia

Nuraini Sibuea, Andi Rio Ramadhan, Ratna Tri Aulia, Siska Khairunnisa, Rado Yendra, Arisman Adnan

Received	Revised	Accepted	Published
15 Oct 2022	16 Nov 2022	29 Nov 2022	15 Dec 2022

Citation :

Nuraini Sibuea, Andi Rio Ramadhan, Ratna Tri Aulia, Siska Khairunnisa, Rado Yendra, Arisman Adnan, "Comparison of Linear Regression Method and Holt-Winters Method in Oil and Gas Export Volume Forecasting in Indonesia," International Journal of Mathematics Trends and Technology (IJMTT), vol. 68, no. 12, pp. 7-12, 2022. Crossref, https://doi.org/10.14445/22315373/IJMTT-V68I12P502

Abstract

The COVID-19 pandemic caused a decline in oil and gas exports from Indonesia in 2019. After some months, the Covid-19 pandemic started to improve slowly. This had an impact on the economy's recovery, which led to an increase in oil and gas exports once more. This article will use the Linear Regression approach and the Holt-Winters method to forecast the amount of oil and gas exported from Indonesia. In particular, the Holt-Winters method is subdivided into the Holt-Winters multiplicative and Holt-Winters additive. Comparing these three approaches, it is evident that the multiplicative Holt-Winters method is superior to the Linear Regression and Holt-Winters additive methods, which approach out-of-sample data for oil and gas volume forecasting in 2018. This is also directly related to the predictive power of the Holt-Winters multiplicative method, which may approach the actual data for six years longer than other methods. The Mean Absolute Error (MAE) of the three approaches indicates that the multiplicative error of the Holt-Winters method is less than that of the other two ways. Therefore, it may be inferred that the Holt-Winters multiplicative technique is superior for predicting Indonesian oil and gas exports.

Keywords

Oil and Gas Exports, Holt-Winters Method, Regression Method, Forecasting.

References

[1] Adam Aboode, “Anomaly Detection in Time Series Data Baed on Holt-Winters Method,” 2018.
[2] Aydin G, “Regression Models for Forecasting Global Oil Production,” Petroleum Science and Technology, vol. 33, pp. 1822– 1828, 2015. Crossref, https://doi.org/10.1080/10916466.2015.1101474
[3] Central Bureau of Statistics (BPS), “Export Volume of Oil and Gas-Non Oil and Gas (Thousand Tonnes), Jakarta,” 2022. [Online]. Available: http://www.bps.go.id/indicator/8/2172/11/volume-ekspor-migas-nonmigas.html
[4] Pudji Djatmiko, and Nugroho SBM, “Impact Non-Oil and Gas Exports and Oil and Gas Exports on the Position of Indonesia Foreign Exchange Reserves,” Journal of Economics & Business Education, vol. 7, no. 8, pp. 87-100, 2019. Crossref, http://doi.org/10.21009/JPEB.007.1.8
[5] Chaido Dritsaki, Dimitrios Niklis, and Pavlos Stamatiou, “Oil Consumption Forecasting using ARIMA Models: An Empirical Study for Greece,” International Journal of Energy Economics and Policy, vol. 11, no. 4, pp. 214-224, 2021. Crossref, https://doi.org/10.32479/ijeep.11231
[6] Hudzaifah Hasri, Siti Armiza Mohd Aris, and Robiah Ahmad, “Linear Regression and Holt’s Winter Algorithm in Forecasting Daily Coronavirus Disease 2019 Cases in Malaysia: Preliminary Studi,” IEEE National Biomedical Engineering Conference (NBEC), Kuala Lumpur, Malaysia, pp. 157-160, 2021. Crossref, https://doi.org/10.1109/NBEC53282.2021.9618763
[7] Indria Mayesti, et al., “Oil and Gas Export Value Analysis and the Economy Indonesia Period 2010-2021,” Budapest International Research and Critics Institute-Journal (BIRCI-Journal), vol. 5, no. 4, pp. 29179-29184, 2022. Crossref, https://doi.org/10.33258/birci v5i4.7069
[8] Douglas C. Montgomery, “Introduction to Time Series Analysis and Forecasting,” 2008.
[9] Sri Endang Rahayu, “Analysis of Indonesian Non-Oil and Gas Exports Development by Manufacturing Sector 2016-2020 Period,” Journal of International Conference Proceedings (JICP), vol. 4, no. 2, pp. 576-583, 2021. Crossref, https://doi.org/10.32535/jicp.v4i2.1294
[10] Rhuan Carlos Martins Ribeiro, Glauber Tadaiesky Marques, and Paulo Cerqueira dos Santos Júnior, “Holt-Winters Forecasting for Brazilian Natural Gas Production,” International Journal of Innovation Education and Research, vol. 7, no. 6, pp. 119-129, 2019. Crossref, https://doi.org/10.31686/ijier.Vol7.Iss6.1559
[11] Niveen Badra, Michael shawky Sabry, and Hatem Abdel-Latif, “Comparison Between Regression and Arima Models in Forecasting Traffic Volume,” Australian Journal of Basic and Applied Sciences, vol. 1, no. 2, pp. 126-136, 2007.
[12] Martin Luter Purba, Meiline Sihombing, and Jusmer Sihotang, “Analysis of the Influence of Oil and Gas Exports, Non-Oil and Gas Exports of Foreign Investment on Indonesia's Economic Growth in 2000-2019,” Journal of Economics and Business, vol. 2, no. 2, 2021. Crossref, https://doi.org/10.36655/jeb.v2i2.557
[13] Anjar Wanto, et al., “Forecasting the Export and Import Volume of Crude Oil, Oil Products and Gas Using ANN,” IOP Conference Series: Journal of Physics: Conference Series, vol. 1255, p. 012016, 2019. Crossref, https://doi.org/10.1088/1742- 6596/1255/1/012016
[14] Cort J. Willmott, and Kenji Matsuura, “Advantages of the Mean Absolute Error (MAE) Over The Root Mean Square Error (RSME) in Assessing Average Model Porformance,” Climate Research, vol. 30, no. 1, pp. 79-82, 2005.