ABSTRACT
The spot commodities market exhibits both extreme volatility and price spikes, which lead to heavy-tailed distributions of price change and autocorrelation. This article uses various Lévy jump models to capture these features in a panel of agricultural commodities observed between January 1990 and February 2014. The results show that Levy jump models outperform the continuous Gaussian model. Our results prove that assuming a constant volatility or even a deterministic volatility and drift structure of agricultural commodity spot prices is not realistic and is less efficient than the stochastic assumption. The findings demonstrate an interesting correlation between volatility and jumps for a given commodity i, but no relationship between the volatility of commodity i and the probability of jumps of commodity j.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Shiller (Citation1981) considers excess volatility to be price movements that are excessive relative to changes in fundamentals, i.e., supply or demand shocks greater than what the efficient market hypothesis would predict.