Bayesian inference in proxy SVARs with incomplete identification: Re-evaluating the validity of monetary policy instruments

• 成果类型: Article
• 署名作者: Nguyen, Lam
• 刊物名称: JOURNAL OF MONETARY ECONOMICS
• ISSN/ISSBN: 0304-3932
• DOI: 10.1016/j.jmoneco.2025.103813
• 发表日期: 2025
• 关键词: Proxy structural vector autoregressions; VAR identification; sign restrictions; instrumental variables; Subjective Bayesian inference; monetary policy

摘要

Instrument validity cannot be tested in a just-identified model, and it is not clear what conclusion to draw when instrument validity is rejected in an over-identified model. In practice, researchers tend to regard instruments as valid when they lead to sensible inferences. This paper develops a proxy structural vector autoregression with imperfect confidence in instrument validity, enabling researchers to incorporate and investigate those prior beliefs alongside other identifying information such as sign restrictions. The empirical application offers a new explanation to the observation in Stock and Watson (2012) that shocks predicted by different monetary policy instruments are correlated with oil and fiscal policy shocks, but not with each other.