Comparing DSGE-VAR forecasting models: How big are the differences?

Andra C. Ghent

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


I generate priors for a vector autoregression (VAR) from a standard real business cycle (RBC) model, an RBC model with capital-adjustment costs and habit formation, and a sticky-price model with an unaccommodating monetary authority. The response of hours worked to a TFP shock differs sharply across these models. I compare the accuracy of forecasts made from each of the resulting dynamic stochastic general equilibrium vector autoregression (DSGE-VAR) models. Despite having different structural characteristics, the DSGE-VARs are comparable in terms of forecasting performance. As in previous work, DSGE-VARs compare favorably with atheoretical VARs.

Original languageEnglish (US)
Pages (from-to)864-882
Number of pages19
JournalJournal of Economic Dynamics and Control
Issue number4
StatePublished - Apr 2009
Externally publishedYes


  • Business cycles
  • Economic fluctuations
  • Hours debate
  • Model evaluation
  • Priors from DSGE models

ASJC Scopus subject areas

  • Economics and Econometrics
  • Control and Optimization
  • Applied Mathematics


Dive into the research topics of 'Comparing DSGE-VAR forecasting models: How big are the differences?'. Together they form a unique fingerprint.

Cite this