Whenever question arises about uncertainty, it can be tackled by Bayesian tools and methods, with the help of priori probabilities and posterior probabilities. In general classical statistics selects just the "best" model and rejects all the others, even of they are only marginally worse than the best model, perhaps the model is a good fit, but in case of forecasting the future; there it fails. Now, the problem is uncertainty about model, in this book detailed discussion about Bayesian analysis and methods, that in contrast, will combine models of highly comparable for forecasting.