- مبلغ: ۸۶,۰۰۰ تومان
- مبلغ: ۹۱,۰۰۰ تومان
We consider the forecasting of macroeconomic variables that are subject to revisions, using Bayesian vintage-based vector autoregressions. The prior incorporates the belief that, after the first few data releases, subsequent ones are likely to consist of revisions that are largely unpredictable. The Bayesian approach allows the joint modelling of the data revisions of more than one variable, while keeping the concomitant increase in parameter estimation uncertainty manageable. Our model provides markedly more accurate forecasts of postrevision values of inflation than do other models in the literature.
In this paper we propose modelling and forecasting macroeconomic variables that are subject to revisions using a Bayesian vintage-based VAR. The vintage-based VAR models in the literature are typically univariate; that is, they model the relationships between different maturities of data of a single variable. While such approaches have shown promise in forecasting revisions of data for which initial estimates have been published, they are less successful at forecasting post-revision values of future observations. The use of a Bayesian approach allows us to build multivariate multiple-vintage models. In our empirical work, these models are estimated for output growth and inflation, and are shown to provide competitive forecasts of the post-revision (or fully-revised) values of future inflation.