Calibrating the size of shocks to match dispersions of key variables

General DSGE Modeling
1

Dear all,

I am currently working on calibrating the size of shocks to match the dispersion of key variables, as well as some other parameters (such as the investment adjustment cost parameter). My primary focus, however, is on calibrating the size of the shocks. Initially, I adopted an approach that involves minimizing an objective function consisting of the squared differences between the target and the second-order moments generated by the model.

While browsing the forum, I came across the GMM/SMM approach.

Could you advise which method might be more suitable for my case? Additionally, where can I find a current-version code sample for implementing this approach?

Any guidance would be greatly appreciated. Thank you!

2

GMM/SMM would be for formal estimation, i.e. if you want to put confidence bounds to your estimates. For simple parameterizations that is often an overkill. You can find some examples at tests/estimation/method_of_moments · master · Dynare / dynare · GitLab