Using the estimated parameters to do a simulation

[AlexLee]

Hi Everyone,

Step one: I calibrated a model, and it worked. Step two: After that, I did an estimation, and it also worked. Step three: I used the estimated posterior mean to replace the calibrated parameters in Step one, and the model did not work. It said that the BK conditions are not satisfied. That is kind of very weird. I could not figure this out. Why does this happen?

Thank you,
Alex

[jpfeifer]

See

Stoch_simul after estimation vs stoch_simul using posterior

For every parameter vector the BK conditions are satisfied in estimation. But the estimation result is none of these parameters. It is not the median, but the mean. The mean is a convex combination of the estimated parameters. If the stability region is a convex set of the parameters, this particular convex-combination called mean will be inside the stability region. However, in some cases the stability region is not a convex set and the mean parameter vector is outside of it. In these cases, s…

[AlexLee]

Thank you Prof. Pfeifer for answering my question. I would like to do a counterfactual analysis. I want to change the value of a calibrated parameter , and use the estimated values for other parameters. However, when I chose the posterior mean to do this, the BK conditions are not satisfied. Based on this fact, Can I still do a counterfactual analysis? In other words, Can I use a different way to do the counterfactual analysis.

[jpfeifer]

See

Stoch_simul after estimation vs stoch_simul using posterior

After running the MCMC, the parameters will be set to the posterior mean. If you want to simulate the model at the posterior mode instead of the mean, you would need xparam = get_posterior_parameters('mode'); stoch_simul(order=1,irf=20);