Robust Supply Chains in the Agri-food Industry

Prices for agricultural products are critical for producers and consumer alike. On the producer side higher prices signal the degree of scarcity of a crop. Given a high price elasticity of supply, accurate and timely price forecasts for the main crops provide Austrian farmers with a valuable signal to shift land use towards more profitable crops, thus supporting the security of food supply in Austria. We apply three classes of time series models to producer prices of four popular crops in Austria: milling wheat, quality wheat, rapeseed, and maize. Using explanatory variables from futures markets and international organisations improves the model performance. The proposed forecasting cycles reflect the dates of decision making in sowing and harvesting each crop. We find that the Mallows Model Averaging method produces the smallest average forecast error, but a combined model forecast ranks best among all alternatives and beats both, individual model forecasts and the prices of matching futures contracts.