Modeling agricultural commodity prices volatility in Sudan using autoregressive conditional heteroskedasticity models

Abstract EN

This paper attempts to model and forecast agricultural commodity prices volatility in the Sudan using Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models.

Various types of GARCH models for instance; GARCH (1,1), EGARCH(1,1), TGARCH(1,1) and APARCH(1,1) were applied to four agricultural commodities cover the period 1st January 1997 – 30th September 2010; to test the hypothesis of persistence, asymmetry and volatility of the prices of agricultural commodities.

It is found that, the sum of ARCH and GARCH coefficients is not close to one in all commodities indicating that volatility shocks is not quite persistent, moreover volatility, changes in volatility are not present in the commodity prices returns series (i.e.

No volatility and changes in volatility in surgham, seaseme, millet and wheat of agricultural commodities), furthermore Asymmetry in volatility is achieved in all commodities except for wheat, in addition none of agricultural commodity has an impact of bad news on the price volatility, while good news on the price volatility is presents in seaseme commodity, lastly leverage effect is not present on the conditional variance.

Therefore (GARCH) models are appropriate in modeling and forecasting agricultural commodity prices volatility.