Mcmc sampling from composite parameters

Estimation
1

Dear all,

I’m trying to estimate the slope of a NKPC slope in the basic NK model (gali). From my understanding I could estimate the individual parameters that make it up then put those in and obtain the value for the slope. But is there a way I can estimate (same use the mcmc to sample) the slope so that I could get the standard error for example for the slope?
Thank you al in advancel.

2

There are two ways you could do this.

  1. Define an auxiliary variable (not parameter) that stores the slope of the NKPC slope and then compute its distribution via moments_varendo
  2. Use the posterior_function-command to sample from the posterior and write a function to compute the NKPC slope in this posterior function