Adaptive Hierarchical Priors for High-Dimensional Vector Autoregessions

Dimitris Korobilis and Davide Pettenuzzo ()

No 115, Working Papers from Brandeis University, Department of Economics and International Business School

Abstract: This paper proposes a scalable and simulation-free estimation algorithm for vector autoregressions (VARs) that allows fast approximate calculation of marginal posterior distributions. We apply the algorithm to derive analytical expressions for popular Bayesian shrinkage priors that admit a hierarchical representation and which would typically require computationally intensive posterior simulation methods. The proposed algorithm is modular, parallelizable, and scales linearly with the number of predictors, allowing fast and efficient estimation of large Bayesian VARs. The benefits of our approach are explored and computational gains of the proposed estimation algorithm and priors. Second, a forecasting exercise involving VARs estimated on macroeconomic data demonstrates the ability of hierarchical shrinkage priors to find useful parsimonious representations. Finally, we show that our approach can be used successfully for structural analysis and can replicate important features of structural shocks predicted by economic theory.

Keywords: Bayesian VAR's; Mixture priors; large datasets; macroeconomic forecasting (search for similar items in EconPapers)
JEL-codes: C11 C13 C32 C53 (search for similar items in EconPapers)
Pages: 53 pages
Date: 2017-09
New Economics Papers: this item is included in nep-for and nep-ore
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://www.brandeis.edu/economics/RePEc/brd/doc/Brandeis_WP115.pdf (application/pdf)

Related works:
Journal Article: Adaptive hierarchical priors for high-dimensional vector autoregressions (2019)
Working Paper: Adaptive Hierarchical Priors for High-Dimensional Vector Autoregressions (2018)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:brd:wpaper:115

Access Statistics for this paper

More papers in Working Papers from Brandeis University, Department of Economics and International Business School Contact information at EDIRC.
Bibliographic data for series maintained by Andrea Luna ().