DYNAMIC EQUILIBRIUM ECONOMIES - A FRAMEWORK FOR COMPARING MODELS AND DATA

We propose a constructive, multivariate framework for assessing agreem ent between (generally misspecified) dynamic equilibrium models and da ta, which enables a complete second-order comparison of the dynamic pr operties of models and data. We use bootstrap algorithms to evaluate t he significance of deviations between models and data, and we use good ness-of-fit criteria to produce estimators that optimize economically- relevant loss functions. We provide a detailed illustrative applicatio n to modelling the U.S. cattle cycle.