Regime-switching stochastic volatility: Evidence from the crude oil market

• Published in: Energy economics Vol. 31; no. 5; pp. 779 - 788
• Main Author: Vo, Minh T.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.09.2009; Elsevier; Elsevier Science Ltd
• Series: Energy Economics
• Subjects: Applied sciences; Bayesian analysis; Crude oil; Crude oil price; Crude oil price Regime switching Stochastic volatility MCMC method Forecasting; Crude oil prices; Economic data; Energy; Energy economics; Exact sciences and technology; Forecasting; Forecasts; Fossil fuels and derived products; General, economic and professional studies; Markov analysis; Markovian processes; MCMC method; Methodology. Modelling; Monte Carlo simulation; Oil; Oil market; Oil price; Regime switching; Stochastic models; Stochastic processes; Stochastic volatility; Studies; Volatility; Crude oil price; C22; G11; Regime switching; MCMC method; G10; G32; Stochastic volatility; Forecasting; Monte Carlo method; Volatility; Stochastic process; Modeling; Petroleum; Markov chain; Price; Bayes forecasting; Oil economy
• ISSN: 0140-9883; 1873-6181
• DOI: 10.1016/j.eneco.2009.05.001

This paper incorporates regime-switching into the stochastic volatility (SV) framework in an attempt to explain the behavior of crude oil prices in order to forecast their volatility. More specifically, it models the volatility of oil return as a stochastic volatility process whose mean is subject to shifts in regime. The shift is governed by a two-state first-order Markov process. The Bayesian Markov Chain Monte Carlo method is used to estimate the models. The main findings are: first, there is clear evidence of regime-switching in the oil market. Ignoring it will lead to a false impression that the volatility is highly persistent and therefore highly predictable. Second, incorporating regime-switching into the SV framework significantly enhances the forecasting power of the SV model. Third, the regime-switching stochastic volatility model does a good job in capturing major events affecting the oil market.