Crude oil price prediction: A comparison between AdaBoost-LSTM and AdaBoost-GRU for improving forecasting performance
- Authors
- Busari, Ganiyu Adewale; Lim, Dong Hoon
- Issue Date
- Dec-2021
- Publisher
- Pergamon Press Ltd.
- Keywords
- Crude Oil Price Forecast; AdaBoost algorithm; Long short-term memory; Gated recurrent unit; Forecasting
- Citation
- Computers & Chemical Engineering, v.155
- Indexed
- SCIE
SCOPUS
- Journal Title
- Computers & Chemical Engineering
- Volume
- 155
- URI
- https://scholarworks.gnu.ac.kr/handle/sw.gnu/2917
- DOI
- 10.1016/j.compchemeng.2021.107513
- ISSN
- 0098-1354
1873-4375
- Abstract
- Crude oil plays an important role in the world economy and contributes to more than one third of energy consumption worldwide. The better forecasting of its fluctuating price is crucial for policymaking. Although various methods had been previously applied for forecasting crude oil price which including both statistical models and Artificial Neural Network models but we find that there is no single paper that compares the AdaBoost-LSTM and AdaBoost-GRU models for improving forecasting performance. We proposed AdaBoost-GRU in which the GRU model was built, put inside sklearn wrapped and finally boost by AdaBoost Regressor. The predictive power of the proposed model was compared with AdaBoost-LSTM using daily Crude oil prices from October 23, 2009, to June 23, 2021, and single LSTM and single GRU were used as the benchmarking models. Forecasting performances were measured using five different metrics namely; the mean absolute error (MAE), the root mean squared error (RMSE), the Scatter Index (SI), the mean absolute percentage error (MAPE), and the weighted mean absolute percentage error (WMAPE). We have demonstrated that the AdaBoost-LSTM and the AdaBoost-GRU models outperform the benchmarking models as expected, and the empirical results show that the AdaBoost-GRU is superior to all models studied in this research. (c) 2021 Elsevier Ltd. All rights reserved.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - 자연과학대학 > Dept. of Information and Statistics > Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.gnu.ac.kr/handle/sw.gnu/2917)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.