Forecasting crude oil risk: A multiscale bidirectional generative adversarial network based approach
作者：
Highlights：
• The multiscale risk structure is decomposed using MEMD model.
• The mixture of HS model is used to model the risk structure.
• The transient risk factor is identified and modeled.
• BiGAN is introduced to extend and augment the limited transient risk data.
• Improvement in risk forecasting accuracy has been achieved.
摘要
•The multiscale risk structure is decomposed using MEMD model.•The mixture of HS model is used to model the risk structure.•The transient risk factor is identified and modeled.•BiGAN is introduced to extend and augment the limited transient risk data.•Improvement in risk forecasting accuracy has been achieved.
论文关键词：Crude oil risk,Value-at-risk,Multivariate empirical mode decomposition (MEMD) model,Multi-scale analysis,Bidirectional generative adversarial network model
论文评审过程：Received 8 June 2022, Revised 20 August 2022, Accepted 29 August 2022, Available online 3 September 2022, Version of Record 19 September 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118743
