COMPUTATION OF BUSINESS CYCLE MODELS: A COMPARISON OF NUMERICAL METHODS

Abstract

We compare the numerical methods that are most widely applied in the computation of the standard business cycle model with flexible labor. The numerical techniques imply economically insignificant differences with regard to business cycle summary statistics. Furthermore, these results are robust with regard to the choice of the functional form of the utility function and the model's parameterization. In addition, the extended path approach, albeit time-consuming, and the Galerkin projection are found to be the most accurate methods, given that we have not used function approximations beyond the second degree.