Confidence Regions for Calibrated Parameters in Computable General Equilibrium Models

To assess the uncertainty of calibrated parameters in computable general equilibrium (CGE), we study two methods for building confidence sets on these parameters. The first method uses a projection technique from (sampling or Bayesian) confidence sets on the free parameters of a deterministic CGE model. The second one allows for random disturbances in the equations of the model and exploits simulation techniques. Applications to CGE models of the Moroccan economy are presented.