Exploiting the monthly data flow in structural forecasting

• 成果类型: Article
• 署名作者: Giannone, Domenico; Monti, Francesca; Reichlin, Lucrezia
• 署名单位: Bank of England; Federal Reserve System - USA; Federal Reserve Bank - New York; University of London; London Business School
• 刊物名称: JOURNAL OF MONETARY ECONOMICS
• ISSN/ISSBN: 0304-3932
• DOI: 10.1016/j.jmoneco.2016.10.011
• 发表日期: 2016
• 页码: 201-215
• 关键词: DSGE models; forecasting; temporal aggregation; Mixed frequency data; Large datasets

摘要

A quarterly stochastic general equilibrium (DSGE) model is combined with a now-casting model designed to read timely monthly information as it becomes available. This implies (1) mapping the structural quarterly DSGE with a monthly version that maintains the same economic restrictions; (2) augmenting the model with a richer data set and (3) updating the estimates of the DSGE's structural shocks in real time following the publication calendar of the data. Our empirical results show that our methodology enhances the predictive accuracy in now-casting. An analysis of the Great Recession also shows that our framework would have helped tracing the DSGE's structural shocks in real time, obtaining, for example, a more timely account of the 2008 contraction. Crown Copyright (C) 2016 Published by Elsevier B.V. All rights reserved.