Partial stochastic analysis with the European Commission's version of the AGLINK-COSIMO model

The Directorate-General for Agriculture and Rural Development (DG AGRI) publishes annually its medium-term agricultural outlook for the main agricultural sectors (i.e. cereals, oilseeds, sugar, meat, dairy, and biofuels) using a partial equilibrium model. The report contains EU-wide projections of supply balance sheets (production, consumption, exports, imports, and change in stocks) for the next 8-10 years. It is inevitable that the results from a partial equilibrium simulation model are conditional on values used for variables that enter the model exogenously. These exogenous variables include some of the key drivers of market behaviour. Because of the uncertainty surrounding their assumed values, it is very useful to conduct sensitivity analysis with respect to key exogenous variables. Stochastic analysis has been used in the DG AGRI agricultural outlooks in both 2011 and 2012 to assess the degree of sensitivity of the baseline projections to uncertainty in the macroeconomy and fluctuations in agricultural yields. This report presents the methodology underlying that analysis.