Modeling and forecasting volatility in global food commodity prices

Abstract

To capture the volatility in the global food commodity prices, we employed two competing models, the thin tailed the normal distribution, and the fat-tailed Student t-distribution models. Results based on wheat, rice, sugar, beef, coffee, and groundnut prices, during the sample period from October 1984 to September 2009, show the t-distribution model outperforms the normal distribution model, suggesting that the normality assumption of residuals which are often taken for granted for its simplicity may lead to unreliable results of the conditional volatility estimates. The paper also shows that the volatility of food commodity prices characterized with the intermediate and short memory behavior, implying that the volatility of food commodity prices is mean reverting.