The Use and Mis-Use of SVARs for Validating DSGE Models

Paul Levine (), Joseph Pearlman, Alessio Volpicella () and Bo Yang
Additional contact information
Bo Yang: Swansea University

No 522, School of Economics Discussion Papers from School of Economics, University of Surrey

Abstract: This paper studies the potential ability of an SVAR to match impulse response functions of a well-established estimated DSGE model. We study the invertibility (fundamentalness) problem setting out conditions for the RE solution of a linearized Gaussian NK-DSGE model to be invertible taking into account the information sets of agents. We then estimate an SVAR by generating artificial data from the theoretical model. A measure of approximate invertibility, where information can be imperfect, is constructed. Based on the VAR(1) representation of the DSGE model, we compare three forms of SVAR-identification restrictions; zero, sign and bounds on the forecast error variance, for mapping the reduced form residuals of the empirical model to the structural shocks of interest. Separating out two reasons why SVARs may not recover the impulse responses to structural shocks of the DGP, namely non-invertibility and inappropriate identification restrictions, is then the main objective of the paper.

JEL-codes: C11 C18 C32 E32 (search for similar items in EconPapers)
Pages: 89 pages
Date: 2022-06
New Economics Papers: this item is included in nep-dge, nep-ecm and nep-mac
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://repec.som.surrey.ac.uk/2022/DP05-22.pdf (application/pdf)

Related works:
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:sur:surrec:0522

Access Statistics for this paper

More papers in School of Economics Discussion Papers from School of Economics, University of Surrey Contact information at EDIRC.
Bibliographic data for series maintained by Ioannis Lazopoulos ().