Vortrag
Assessing Identifying Restrictions in SVAR Models

Michele Piffer


IAAE 2016 Annual Conference
Mailand, Italien, 22.06.2016 - 25.06.2016



Abstract:
This paper proposes a Bayesian approach to assess if the data support candidate set-identifying restrictions in Vector Autoregressive models. I study the case of sign restrictions. The researcher expresses her uncertainty regarding the validity of the restrictions using a prior distribution that covers the parameter space both where the restrictions are satisfied and where they are not satisfied. The correlation in the data then determines whether the probability mass in favour of the restrictions increases or not from prior to posterior. I apply the proposed Bayesian assessment to a two-equation model of labour demand and supply, and to a New Keynesian model.

Abstract

This paper proposes a Bayesian approach to assess if the data support candidate set-identifying restrictions in Vector Autoregressive models. I study the case of sign restrictions. The researcher expresses her uncertainty regarding the validity of the restrictions using a prior distribution that covers the parameter space both where the restrictions are satisfied and where they are not satisfied. The correlation in the data then determines whether the probability mass in favour of the restrictions increases or not from prior to posterior. I apply the proposed Bayesian assessment to a two-equation model of labour demand and supply, and to a New Keynesian model.

Themen: Surveymethodologie und Data-Science



JEL-Classification: C32;C11
Keywords: Identification, Bayesian Econometrics, Sign Restrictions
DIW-Link
Array

keyboard_arrow_up