Leverage effects and stochastic volatility in spot oil returns: A Bayesian approach with VaR and CVaR applications

Liyuan Chen, Paola Z Zerilli, Christopher Baum

Research output: Contribution to journalArticlepeer-review


Crude oil markets have been quite volatile and risky in the past few decades due to the large fluctuations of oil prices. A large number of studies have shown that oil price fluctuations could have considerable impact on economic activities.
We contribute to the current debate by testing for the existence of the leverage effect when considering daily spot returns in the WTI and Brent crude oil markets and by studying the direct impact of the leverage effect on measures of risk such as VaR and CVaR. More specifically, in order to address the risk faced by oil suppliers and oil consumers we model spot crude oil returns using Stochastic Volatility (SV) models with various distributions of the errors. Among other cases, we test the assumption of Asymmetric Laplace Distributed
(ALD) errors in order to more carefully model the type of risk faced by oil suppliers versus the risk faced by oil buyers. We find that the introduction of the leverage effect in the traditional SV model with Normally distributed errors is capable of adequately estimating risk (in a VaR and CVaR sense) for conservative oil suppliers in both the WTI and Brent markets while it tends to overestimate risk for more speculative oil suppliers. In comparison, the assumption of ALD errors leads to overestimating risk for both types of investors.
Original languageEnglish
Pages (from-to)111-129
Number of pages19
JournalEnergy economics
Early online date18 Apr 2018
Publication statusPublished - Mar 2019

Bibliographical note

© 2018 Elsevier B.V. All rights reserved. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.


  • Asymmetric Laplace distribution
  • Bayesian Markov chain Monte Carlo
  • Conditional value-at-risk
  • Leverage effect
  • Stochastic volatility model
  • Value-at-risk

Cite this