Revisiting value-at-risk and expected shortfall in oil markets under structural breaks: The role of fat-tailed distributions

Modeling the volatility of oil prices is extremely crucial from a risk management perspective. Value-at-risk (VaR) and Expected Shortfall (ES), the two most popular measures of risk in financial markets, are dependent on the volatility of the oil prices. In this study, Pearson Type-IV and Johnsons Su distributions are explored as two alternate distributions with characteristics, such as asymmetry and heavy tail, to model the volatility and forecast VaR/ES. The estimation is carried out under endogenously determined structural breaks from the data. Various Backtesting methodologies are employed to test the efficacy of the forecasts. The empirical results obtained show that the models with Pearson's Type-IV and Johnson's Su distributions outperform other fat-tailed distributions and the normal distribution especially at the 1% (for long positions) and 99% (for short positions) level. This has policy implications for the oil producing companies, market participants, regulators in the energy sector and the government in general.

Automated summary

Modeling the volatility of oil prices is crucial for risk management, and using alternative distributions like Pearson's Type-IV and Johnson's Su distributions can improve forecasts of Value-at-Risk and Expected Shortfall. Empirical results show that these models outperform other fat-tailed distributions and normal distribution, especially at extreme levels.