Copula stochastic volatility in oil returns: Approximate Bayesian computation with volatility prediction

Modeling the volatility of energy commodity returns has become a topic of increased interest in recent years, because of the important role it plays in today's economy. In this paper we propose a novel copula-based stochastic volatility model for energy commodity returns that allows for asymmetric volatility persistence. We employ Approximate Bayesian Computation (ABC), a powerful tool to make inferences and predictions for such highly-nonlinear model. We carry out two simulation studies to illustrate that ABC is an appropriate alternative to standard MCMC-based methods when the state transition process is challenging to implement. Finally, we model the volatility of WTI and Brent oil futures' returns with the proposed copula-based stochastic volatility model and show that such model outperforms symmetric alternatives in terms of in- and out-of-sample volatility prediction accuracy.