Hybridized LSTM-GRU model for forecasting the Prices of Crude oil

  • Afeez A. Soladoye Federal university Oye-Ekiti
  • Mutiu B. Falade
  • Charity S. Odeyemi
  • Abubakar Barkindo


Fluctuation in crude oil price doesn’t only affect production or transportation but has secondary effect on all ways of life, as increase cost of production would surely lead to increase in the price of produce and increased in cost of transportation, which would in turn increase cost of living, increase poverty and hunger. Owing to this, accurate forecasting of crude oil price through the application of effective and technological drive approach would help the policy makers, industrial decision makers and investors to make informed and strategic policies that would help the government and stakeholders to prevent unplanned economic risk and damage. Conventionally, crude oil price forecasting is being done using traditional methods like Liner regression, generalized autoregressive conditional heteroskedasticity (GARCH) and autoregressive integrated moving average (ARIMA) models, this model most time struggles with the non-linearity and fluctuating nature of the crude oil price. Moreover, this models require additional manual feature selection and extraction which might be time consuming and computationally exhausting. The acceptance and proven accurate performance of deep learning techniques prompted this study to comparatively employ the major variants of Recurrent Neural Networks (RNN) namely Long-short term memory (LSTM) and Gated Recurrent Units (GRU) for forecasting of crude oil price owing to their strengths. The famous Brent crude oil price dataset was used for forecasting of crude oil price, while this dataset went through series of preprocessing like data and time conversion, data windowing, transformation, variable selection after which LSTM-GRU model was used for forecasting, this hybridized model gave a better performance when evaluated with Mean Absolute Error of 0.014 which outperformed LSTM and GRU and when compared with existing studies. This shows that hybridized LSTM-GRU is a good model for forecasting of crude oil price as the model leverage the strength of LSTM and GRU on sequential data.


Chadha, G., Panambilly, A., Schwung, A. and Ding, S. (2020) “Bidirectional Deep Recurrent Neural Networks for Process Fault Classification” ISA Transactions, 106: pp. 330–342

Cheng, M., Chu, C. and Hsu, H. (2021) “A. study of univariate forecasting methods for crude oil price” Maritime Business Review, 8(1), pp. 32-47

Choi, Y, Park,S., Jun, Pyo, C., Cho, K., Lee H. and Yu., J. (2021) “Deep Learning-Based Stroke Disease Prediction System Using Real-Time Bio Signal” Sensors vol. 21(269) https://doi.org/10.3390/ s21134269

Chung, J.; Gulcehre, C.; Cho, K.; Bengio, Y. (2014). “Empirical evaluation of gated recurrent neural networks on sequence modeling”. arXiv , arXiv:1412.3555

Daneshvar, A., Ebrahimi, M., Salahi, F., Rahmaty, M. and Homayounfar, M. (2022). “Brent Crude Oil Price Forecast Utilizing Deep Neural Network Architectures” Computational Intelligence and Neuroscience

Deng, C.; Ma, L.; Zeng, T. (2021). “Crude Oil Price Forecast Based on Deep Transfer Learning: Shanghai Crude Oil as an Example”. Sustainability 13, 13770. https:// doi.org/10.3390/su132413770

Drachal, K. (2023). “Forecasting the Crude Oil Spot Price with Bayesian Symbolic Regression”. Energies, 16(4). https://doi.org/10.3390/ en1601000

Dutta, S., Mandal, J.K., Kim, T.H. and Bandyopadhyay, S.K. (2020) “Breast Cancer Prediction Using Stacked GRU-LSTM-BRNN” Applied Computer Systems, 25(2), pp. 163–171

Fauzi, F., Wijaya, D. R. and Utami, T. W. (2023) “Brent Crude Oil Price Forecasting using the Cascade Forward Neural Network” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), 7(4), pp. 964-969

Hasan, M., Abedin, M. Z., Hajek, P., Coussement, K., Sultan, M and Lucey, B. (2024) “A blending ensemble learning model for crude oil price forecasting” Annals of Operations Research https://doi.org/10.1007/s10479-023-05810-8.

Lu, Q., Sun, S., Duan, H. and Wang, S. (2021) “Analysis and forecasting of crude oil price based on the variable selection-LSTM integrated model” Energy Informatics, 4(Suppl 2):47

Xian, L. J., Ismail, S., Mustapha, A., Abd Wahab, M. H., Syed Idrus, S. (2020). “Crude Oil Price Forecasting Using Hybrid Support Vector Machine” IOP Conference Series: Materials Science and Engineering 917, 012045

Zhang, H. and Hong, M. (2022). “Forecasting crude oil price using LSTM neural networks”. Data science in Finance and Economics, 2(3), pp. 163-180