Dynare forums

Dear all,

I'm interested in the unconditional first two moments of endogenous variables in an estimated model.

If the theoretical moments are used, the first moments usually come from the calibrated parameters determining the steady states e.g., steady-state technology growth, inflation targets, discount rates etc., so there is no uncertainty around the 1st theoretical moment.

However - and please correct me if I'm wrong - the 2nd theoretical moment can be (and often is) affected by the estimated parameters, and thus the 2nd theoretical moment of endogenous variables has a distribution around it, but Dynare does not report it.

So, my question is, would it be possible to calculate the HPD interval for the 2nd theoretical moment, or should I better go for a simulated moment, if the confidence interval for the unconditional moments are required (e.g. by a journal referee)?

gin

First, a correction: in an estimated model even the first moments can be associated with parameter uncertainty if you are estimating parameters that affect the steady states.

Dynare by default does not provide the distribution of the higher order moments, i.e. there are no HPDIs for variances. You can retrieve them in Dynare, but this is a bit more involved. I am working on making such a rountine available, but it is still weeks to months away.