Simulation-based selection of prediction models

We target an assessment of the potential benfits of basing model selection decisions in a forecasting context on simulations that fuse data information and the structure hypothesized by tentative rival models. These procedures can be applied to any empirical forecasting problems. Our main focus is, however, on macro-economic forecasting. Our 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.