NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Prior Selection for Vector Autoregressions

Domenico Giannone, Michele Lenza, Giorgio E. Primiceri

NBER Working Paper No. 18467
Issued in October 2012
NBER Program(s):   EFG   ME   TWP

Vector autoregressions (VARs) are flexible time series models that can capture complex dynamic interrelationships among macroeconomic variables. However, their dense parameterization leads to unstable inference and inaccurate out-of-sample forecasts, particularly for models with many variables. A solution to this problem is to use informative priors, in order to shrink the richly parameterized unrestricted model towards a parsimonious naìˆve benchmark, and thus reduce estimation uncertainty. This paper studies the optimal choice of the informativeness of these priors, which we treat as additional parameters, in the spirit of hierarchical modeling. This approach is theoretically grounded, easy to implement, and greatly reduces the number and importance of subjective choices in the setting of the prior. Moreover, it performs very well both in terms of out-of-sample forecasting--as well as factor models--and accuracy in the estimation of impulse response functions.

download in pdf format
   (307 K)

email paper

This paper is available as PDF (307 K) or via email.

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w18467

Published: Domenico Giannone & Michele Lenza & Giorgio E. Primiceri, 2015. "Prior Selection for Vector Autoregressions," The Review of Economics and Statistics, MIT Press, vol. 2(97), pages 436-451, May. citation courtesy of

Users who downloaded this paper also downloaded these:
Forbes w18465 The "Big C": Identifying Contagion
Bernanke, Boivin, and Eliasz w10220 Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach
Greenwood and Shleifer w18686 Expectations of Returns and Expected Returns
Jaimovich and Rebelo w18473 Non-linear Effects of Taxation on Growth
Stock and Watson w11467 Implications of Dynamic Factor Models for VAR Analysis
 
Publications
Activities
Meetings
NBER Videos
Themes
Data
People
About

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us