A deep learning ensemble approach for crude oil price forecasting

As crude oil price is influenced by numerous factors, capturing its behavior precisely is quite challenging, and thus leads to the difficulty of forecasting. In this study, a deep learning ensemble approach is proposed to deal with this problem. In our approach, two techniques are utilized. One is a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Energy economics Jg. 66; S. 9 - 16
Hauptverfasser: Zhao, Yang, Li, Jianping, Yu, Lean
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Kidlington Elsevier B.V 01.08.2017
Elsevier Science Ltd
Schlagworte:
Artificial neural networks
Bagging
Cognitive style
Crude oil
Crude oil price forecasting
Crude oil prices
Data analysis
Datasets
Deep learning
Denoising
Energy economics
Ensemble learning
Forecasting
Mathematical models
Multivariate forecasting
Networks
Neural networks
Noise reduction
Petroleum
Stacked denoising autoencoder
Studies
Training
C53
Deep learning
C52
C63
Crude oil price forecasting
C45
Q47
Stacked denoising autoencoder
E37
Ensemble learning
Bagging
Multivariate forecasting
ISSN:0140-9883, 1873-6181
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:As crude oil price is influenced by numerous factors, capturing its behavior precisely is quite challenging, and thus leads to the difficulty of forecasting. In this study, a deep learning ensemble approach is proposed to deal with this problem. In our approach, two techniques are utilized. One is an advanced deep neural network model named stacked denoising autoencoders (SDAE) which is used to model the nonlinear and complex relationships of oil price with its factors. The other is a powerful ensemble method named bootstrap aggregation (bagging) which generates multiple data sets for training a set of base models (SDAEs). Our approach combines the merits of these two techniques and is especially suitable for oil price forecasting. In the empirical study, the WTI crude oil price series are investigated and 198 economic series are used as exogenous variables. Our approach is tested against some competing approaches and shows superior forecasting ability that is statistically proved by three tests. •A deep learning and ensemble learning based forecasting approach is proposed.•198 exogenous variables are included.•Stacked denoising autoencoders (SDAE) is used to model and forecast oil price.•Bootstrapping aggregation (bagging) is used to grow an ensemble.•Effectiveness of the proposed model is statistically confirmed.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0140-9883
1873-6181
DOI:10.1016/j.eneco.2017.05.023