Title: Simulation-based selection of prediction models
Authors: Robert Kunst - Institute for Advanced Studies (Austria) [presenting]
Abstract: The benefits of basing model selection decisions in a forecasting context are assessed on simulations that fuse data information and the structure hypothesized by tentative rival models. These procedures can be applied to any empirical forecasting problems. The focus is, however, on macro-economic applications. The suggested procedure aims at choosing among a small number of tentative forecast models in the presence of data. From models fitted to the data, pseudo-data are generated. Again, the models are applied to the pseudo-data and their out-of-sample performance is evaluated. The ultimate choice of the forecasting model is based on the relative performance of rival models in predicting ``their own data'' and those of the rival model. The project covers the three aspects of a rigorous statistical foundation, of a Monte Carlo evaluation of the procedure, and of exemplary empirical implementations of the method in typical macro-economic applications.