Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review
We propose a new methodology for ranking in probability the commonly proposed drivers of inflation in the new Keynesian model. The approach is based on Bayesian model selection among restricted vector autoregressive (VAR) models, each of which embodies only one or none of the candidate variables as the driver. Simulation experiments suggest that our procedure is superior to the previously used conventional pairwise Granger causality tests in detecting the true driver. Empirical results lend little support to labour share, output gap or unemployment rate as the driver of US inflation.
Original language | English |
---|---|
Journal | Oxford Bulletin of Economics and Statistics |
Volume | 76 |
Issue | 5 |
Pages (from-to) | 715–726 |
ISSN | 0305-9049 |
DOIs | |
Publication status | Published - 2014 |
See relations at Aarhus University Citationformats
ID: 85178923