Forecasting GDP with global components: This time is different

详细信息    查看全文

• 作者：Hilde C. Bjø ; rnland^a ; ^{hilde.c.bjornland@bi.no}Author Vitae ; Francesco Ravazzolo^b ; ^a ; ^{francesco.ravazzolo@unibz.it}Author Vitae ; Leif Anders Thorsrud^a ; ^{leif.a.thorsrud@bi.no}Author Vitae
• 关键词：Bayesian Dynamic Factor Model (BDFM) ; Forecasting ; Model uncertainty ; Global factors
• 刊名：International Journal of Forecasting
• 出版年：2017
• 出版时间：January-March 2017
• 年：2017
• 卷：33
• 期：1
• 页码：153-173
• 全文大小：885 K
• 卷排序：33

文摘

We examine whether a knowledge of in-sample co-movement across countries can be used in a more systematic way to improve the forecast accuracy at the national level. In particular, we ask whether a model with common international business cycle factors adds marginal predictive power over a domestic alternative. We answer this question using a dynamic factor model (DFM), and run an out-of-sample forecasting experiment. Our results show that exploiting the informational content in a common global business cycle factor improves the forecast accuracy in terms of both point and density forecast evaluation across a large panel of countries. We also document evidence showing that the Great Recession has a huge impact on this result, causing a clear shift in preferences towards the model that includes a common global factor. However, this time is different in other respects too, as the performance of the DFM deteriorates substantially for longer forecasting horizons in the aftermath of the Great Recession.