How to implement posterior-based Wald-type test after Dynare Bayesian estimation? Reference: Liu et al. (J Econometrics)

Dynare help (legacy posts)
1

Hello all,

I have estimated a DSGE model using Dynare’s Bayesian estimation framework. After estimation, I obtained the posterior mean vector and posterior covariance matrix of the parameters from oo_.posterior.metropolis.mean and oo_.posterior.metropolis.Variance.

I want to conduct a Wald-type hypothesis test on parameter restrictions, specifically testing linear restrictions like equality between two parameters, e.g., mbarh=mbarf.

I came across the paper: Xiaobin Liu, Yong Li, Jun Yu, Tao Zeng, “Posterior-based Wald-type statistics for hypothesis testing,” Journal of Econometrics

This paper proposes a Wald-type statistic constructed directly from posterior mean and covariance matrices to test such hypotheses.

My questions are:

  1. What is the recommended practical way to implement this test using Dynare output?
  2. Are there any subtleties or adjustments I need to be aware of when applying this posterior-based Wald test?
  3. Could you share example code or best practices for coding this test after Dynare Bayesian estimation?

Thanks in advance for your advice!
Best regards,
Jiacheng