LSTM ve Facebook Prophet Kullanarak Brent Ham Petrol Trendinin Tahmini

Abstract

Ham petrol ve petrol ürünleri, endüstriyel üretimin önemli girdileri arasında olduğu kadar lojistik ve taşımacılıkta da kritik bir rol oynamaktadır. Dolayısıyla, petrol fiyatlarındaki ani artışlar ve düşüşler küresel ekonomilerde ve dahası ekonomiler üzerinde doğrudan veya dolaylı bir etkisi vardır. Ayrıca, gelişmekte olan ekonomilerdeki krizler, büyük ekonomiler arasındaki ticaret anlaşmazlıkları ve petrol fiyatının dinamik doğası, petrol arz ve talebi üzerinde etkisi olmaktadır ve petrol fiyatında zaman zaman oynaklık çok sert olmaktadır. Petrol fiyatlarındaki bu belirsizlikler hem tüketicilere hem de üreticilere ağır potansiyel kayıplar yaratabilmektedir. Bu hızlı değişkenlik ve dalgalanma nedeniyle petrol fiyatlarının tahmin edilmesi küresel öneme sahiptir. Bu çalışmada, Brent Petrol fiyatlarının gelecekteki trendini tahmin edilebilmek için geçmiş değerleri girdi alan Uzun Kısa Süreli Bellek (LSTM) ve Facebook Prophet (FBPr) yöntemleri kullanılmıştır. İki modelin petrol fiyatları için Haziran 1988 ile Haziran 2020 arasında haftalık 32 yıllık veri seti kullanılarak karşılaştırılmış ve en uygun model belirlenmiştir. Veri seti eğitim ve test setleri olmak üzere iki gruba ayrılmıştır; eğitim seti için ilk yirmi beş yıl seçilirken ve son yedi yıl ise tahmin doğruluğunu onaylamak için kullanılmıştır. LSTM ve FBPr modelleri için katsayı tayini (R2) eğitim aşamasında 0.92, 0.89 ve test aşamasında 0.89, 0.62 bulunmuştur. Elde edilen sonuçlar incelendiğinde, LSTM modelinin petrol fiyatlarındaki trendi tahmin etmek için daha iyi sonuç verdiği görülmüştür.

Keywords

• Brent Petrol
• Tahmin Etme
• Derin Öğrenme
• LSTM
• Facebook Prophet

The Prediction of Brent Crude Oil Trend Using LSTM and Facebook Prophet

Abstract

Crude oil and petroleum products are among the critical inputs of industrial production as well as they are having an important role in logistics and transportation. Hence, sudden increases and decreases in oil prices cause particular problems in global economies and thus; they have a direct or indirect effect on economies. Furthermore, due to crises in developing economies, trade disputes between major economies, and the dynamic nature of the oil price effect on demand and supply for oil and petroleum products, and time to time volatility in the oil price are very severe. The uncertainty in oil prices can leave both consumers and producers with heavy potential losses. Due to this rapid variability, predicting oil prices has global importance. In this study, to increase the accuracy and stability, the Long-Short Term Memory (LSTM) and Facebook's Prophet (FBPr) were applied to foresee future tendencies in Brent oil prices considering their previous prices. A comparison of the two models made using the 32-year data set between June 1988 and June 2020 weekly for oil prices, and the model with the best fit was determined. The dataset was split into two sets which are training and test sets—the first twenty-five years used for the training set and the last seven years validating forecasting accuracy. The coefficient determination (R2) for the LSTM and FBPr models found as 0.92, 0.89 in the training stage, and 0.89, 0.62 in the testing stage, respectively. According to the results obtained, the LSTM model has superior results to predict the trend of oil prices.

Keywords

• Brent Oil
• Forecasting
• Deep Learning

References

1. Guo J. Oil price forecast using deep learning and ARIMA. Proc - 2019 Int Conf Mach Learn Big Data Bus Intell MLBDBI 2019 2019:241–7. https://doi.org/10.1109/MLBDBI48998.2019.00054.
2. Chiroma H, Abdulkareem S, Herawan T. Evolutionary Neural Network model for West Texas Intermediate crude oil price prediction. Appl Energy 2015;142:266–73. https://doi.org/10.1016/j.apenergy.2014.12.045.
3. Duan H, Lei GR, Shao K. Forecasting crude oil consumption in China using a grey prediction model with an optimal fractional-order accumulating operator. Complexity 2018;2018. https://doi.org/10.1155/2018/3869619.
4. Wang J, Lei C, Guo M. Daily natural gas price forecasting by a weighted hybrid data-driven model. J Pet Sci Eng 2020;192:107240. https://doi.org/10.1016/j.petrol.2020.107240.
5. Bristone M, Prasad R, Abubakar AA. CPPCNDL: Crude oil price prediction using complex network and deep learning algorithms. Petroleum 2020:1–9. https://doi.org/10.1016/j.petlm.2019.11.009.
6. Dées S, Karadeloglou P, Kaufmann RK, Sánchez M. Modelling the world oil market: Assessment of a quarterly econometric model. Energy Policy 2007;35:178–91. https://doi.org/10.1016/j.enpol.2005.10.017.
7. Cabedo JD, Moya I. Estimating oil price “Value at Risk” using the historical simulation approach. Energy Econ 2003;25:239–53. https://doi.org/10.1016/S0140-9883(02)00111-1.
8. Salvi H, Avdhi Shah, Manthan Mehta, Stevina Correia. Long Short-Term Model for Brent Oil Price Forecasting. Int J Res Appl Sci Eng Technol 2019;7:315–9. https://doi.org/10.22214/ijraset.2019.11050.

9. An J, Mikhaylov A, Moiseev N. Oil price predictors: Machine learning approach. Int J Energy Econ Policy 2019;9:1–6. https://doi.org/10.32479/ijeep.7597.
10. Khashman A, Nwulu NI. Intelligent prediction of crude oil price using Support Vector Machines. 9th IEEE Int Symp Appl Mach Intell Informatics, SAMI 2011 - Proc 2011:165–9. https://doi.org/10.1109/SAMI.2011.5738868.
11. Gabralla LA, Jammazi R, Abraham A. Oil price prediction using ensemble machine learning. Proc - 2013 Int Conf Comput Electr Electron Eng ’Research Makes a Differ ICCEEE 2013 2013:674–9. https://doi.org/10.1109/ICCEEE.2013.6634021.
12. Ishaq MF. DATA MINING FORCASTING Oil and Gas Development Company Ltd . Share Prices Using Orange . 2020.
13. Abdullah SN, Zeng X. Machine learning approach for crude oil price prediction with Artificial Neural Networks-Quantitative (ANN-Q) model. Proc Int Jt Conf Neural Networks 2010;44. https://doi.org/10.1109/IJCNN.2010.5596602.
14. Olofin SO, Oloko TF, Isah KO, Ogbonna AE. Crude oil price–shale oil production nexus: a predictability analysis. Int J Energy Sect Manag 2020;14:729–44. https://doi.org/10.1108/IJESM-05-2019-0004.
15. Gupta N, Nigam S. Crude Oil Price Prediction using Artificial Neural Network. Procedia Comput Sci 2020;170:642–7. https://doi.org/10.1016/j.procs.2020.03.136.
16. Abdollahi H, Ebrahimi SB. A new hybrid model for forecasting Brent crude oil price. Energy 2020;200:117520. https://doi.org/10.1016/j.energy.2020.117520.
17. Latifoglu L, Nuralan KB. Tekil Spektrum Analizi ve Uzun-Kısa Süreli Bellek Ağları ile Nehir Akım Tahmini. Eur J Sci Technol 2020:376–81. https://doi.org/10.31590/ejosat.araconf49.
18. Oğuz K, Pekin MA. Yapay Sinir Ağları ile Esenboğa Havaalanı için Sis Görüş Mesafesinin Tahmin Edilebilirliği. Eur J Sci Technol 2019:542–51. https://doi.org/10.31590/ejosat.452598.
19. Gultepe Y. Makine Öğrenmesi Algoritmaları ile Hava Kirliliği Tahmini Üzerine Karşılaştırmalı Bir Değerlendirme. Eur J Sci Technol 2019:8–15. https://doi.org/10.31590/ejosat.530347.
20. Alpay Ö. LSTM Mimarisi Kullanarak USD/TRY Fiyat Tahmini. Eur J Sci Technol 2020:452–6. https://doi.org/10.31590/ejosat.araconf59.
21. Kızılöz HE. Bilimsel Makalelerin Atıf Sayısı Tahmini. Eur J Sci Technol 2020:370–5. https://doi.org/10.31590/ejosat.araconf48.
22. Aguilera H, Guardiola-Albert C, Naranjo-Fernández N, Kohfahl C. Towards flexible groundwater-level prediction for adaptive water management: using Facebook’s Prophet forecasting approach. Hydrol Sci J 2019;64:1504–18. https://doi.org/10.1080/02626667.2019.1651933.
23. Weytjens H, Lohmann E, Kleinsteuber M. Cash flow prediction: MLP and LSTM compared to ARIMA and Prophet. Electron Commer Res 2019. https://doi.org/10.1007/s10660-019-09362-7.
24. Duarte D, Faerman J. Comparison of Time Series Prediction of Healthcare Emergency Department Indicators with ARIMA and Prophet 2019:123–33. https://doi.org/10.5121/csit.2019.91810.
25. Žunić E, Korjenić K, Hodžić K, Đonko D. Application of Facebook’s Prophet Algorithm for Successful Sales Forecasting Based on Real-world Data. Int J Comput Sci Inf Technol 2020;12:23–36. https://doi.org/10.5121/ijcsit.2020.12203.
26. Samal KKR, Babu KS, Das SK, Acharaya A. Time series based air pollution forecasting using SARIMA and prophet model. ACM Int Conf Proceeding Ser 2019:80–5. https://doi.org/10.1145/3355402.3355417.
27. Borowik G, Wawrzyniak ZM, Cichosz P. Time series analysis for crime forecasting. 26th Int Conf Syst Eng ICSEng 2018 - Proc 2019. https://doi.org/10.1109/ICSENG.2018.8638179.
28. Phutela N, Arushi G, Gupta S, Gabrani G. Forecasting the Stability of COVID-19 on Indian Dataset with Prophet Logistic Growth Model. Infect Dis (Auckl) 2020:1–9. https://doi.org/10.21203/rs.3.rs-32472/v1.
29. NASDAQ. NASDAQ 2020. https://www.nasdaq.com/.
30. Rahman MM, Ghasemi Y, Suley E, Zhou Y, Wang S, Rogers J. Machine Learning Based Computer Aided Diagnosis of Breast Cancer Utilizing Anthropometric and Clinical Features. Irbm 2020. https://doi.org/10.1016/j.irbm.2020.05.005.
31. Medium. Medium 2020. https://medium.com/mlreview/understanding-lstm-and-its-diagrams-37e2f46f1714.
32. Fang WX, Lan PC, Lin WR, Chang HC, Chang HY, Wang YH. Combine Facebook Prophet and LSTM with BPNN Forecasting financial markets: The Morgan Taiwan Index. Proc - 2019 Int Symp Intell Signal Process Commun Syst ISPACS 2019 2019:0–1. https://doi.org/10.1109/ISPACS48206.2019.8986377.
33. Becerra M, Jerez A, Aballay B, Garcés HO, Fuentes A. Forecasting emergency admissions due to respiratory diseases in high variability scenarios using time series: A case study in Chile. Sci Total Environ 2020;706. https://doi.org/10.1016/j.scitotenv.2019.134978.
34. Zheng Y, Zhang L, Zhang X, Wang K, Zheng Y. Forecast Model Analysis for the Morbidity of Tuberculosis in Xinjiang , China 2015:1–13. https://doi.org/10.1371/journal.pone.0116832.