Forecasting using relative entropy

The paper describes a relative entropy procedure for imposing restrictions on simulated forecast distributions from a variety of models. Starting from an empirical forecast distribution for some variables of interest, the technique generates a new empirical distribution that satisfies a set of moment restrictions not used in the construction of the original. The new distribution is informationally as close as possible to the original in the sense of minimizing the Kullback-Leibler Information Criterion, or relative entropy. We illustrate the technique with an example related to monetary policy that shows how to introduce restrictions from economic theory into a model's forecasts.