Dynamics behind agricultural food commodity prices: a forecast.

This paper develops a model that is able to forecast the prices of futures of the agricultural food commodities wheat, corn, soybeans and rice. Using the Diebold Mariano (1995) test of equal predictive ability, it is found that the forecast model provides better forecasts than the random walk, making it perform better than the most important benchmark to beat. The model contains variables that can be placed in four groups: cost of production, change of demand, speculation and change in exchange rate. While analysing the forecast model it is also seen that the price of fertilizer, production of biofuel, speculation intensity and the trade weighted US dollar exchange rate affect one or more of the agricultural food commodity prices negatively. The crude oil price and worldwide GDP growth, on the other hand, impact one or more of the agricultural food commodity prices positively.