Explaining the convenience yield in the WTI crude oil market using realized volatility and jumps

In this paper, we first provide empirical evidence of the existence of intraday jumps in the crude oil price series. We then show that these jumps, in conjunction with realized volatility measures, are important in modeling the convenience yield over the 2001-2010 period. Our empirical results indicate that lagged jumpmean only explains around 16% of the weekly convenience yield. Our best specification, including variation in inventories, eight-week realized variance and the 250-day jumpmean is able to explain around 61% of the weekly convenience yield. Importantly, our results are not driven by the simultaneous determination of the various variables at work as we only use lagged variables in all regressions.