Hedge Performance of Oil Futures Contracts Using Mixed Normal Dynamic Conditional Correlation Models
Jong-Ha Weon ․Sang-Kuck Chung

― Abstract―

Bivariate mixed normal time varying GARCH models are proposed for capturing the skewness and kurtosis detected in both conditional and unconditional return distributions. Overall, this study investigates separating the effects of positive and negative basis on market volatility, and the correlation between two markets, as well as jointly incorporating the long memory effect of the basis on market returns. This not only provides better descriptions of the dynamic behavior of the oil prices, but also plays an important role in discovering dynamic hedging strategies. The proposed models are compared with different standard bivariate GARCH models in terms of both the percentage variance reduction of the out-of-sample hedged portfolio and the statistical significance test of the performance improvements. All models are applied to the crude and heating oil markets, and the out-of-sample evaluation is carried out by comparing the hedged portfolio variances over hedge horizons ranging from one to 50 days.

Keywords : Dynamic Hedge Performances, Regime-dependent Conditional Correlations, Bivariate Mixed Normal DCC GARCH, SPA tests

JEL Classification Number : C13, C32, G13