# Bayesian IRF

**URL:** <https://forum.dynare.org/t/bayesian-irf/1594>\
**Category:** ML/Bayesian estimation\
**Created:** [11 August 2010 14:40 UTC](https://forum.dynare.org/t/bayesian-irf/1594 "2010-08-11T14:40:46Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![rain](https://forum.dynare.org/letter_avatar_proxy/v4/letter/r/e95f7d/32.png) [@rain](https://forum.dynare.org/u/rain)\
**Post date:** [11 August 2010 14:40 UTC](https://forum.dynare.org/t/bayesian-irf/1594/1 "2010-08-11T14:40:46Z")

</div>

Hi everyone,

I have a question about the “bayesian irf” command (after MCMC) and “stoch\_simul” command after MCMC. I compare the impulse responses produced by both command (the mean of the bayesian irf v.s. stoch\_simul irf) but they are not same. Could someone please tell me what is the source of this difference?

Thank you very much in advance.

---

<div class="post-metadata">

**Author:** ![MichelJuillard](https://forum.dynare.org/user_avatar/forum.dynare.org/micheljuillard/32/5318_2.png) [@MichelJuillard](https://forum.dynare.org/u/MichelJuillard)\
**Post date:** [11 August 2010 14:53 UTC](https://forum.dynare.org/t/bayesian-irf/1594/2 "2010-08-11T14:53:27Z")

</div>

The mean of the the Bayesian IRFs is the mean of the distribution of the IRFs generated when parameters are drawn from the posterior distribution.

When you run stoch\_simul after MCMC, the parameters of the model are set to their mean value in the posterior distribution.

But an IRF is a nonlinear function of the parameters, therefore the mean of the IRF is not identical to the IRF for the mean value of the parameters

Best

Michel

---

<div class="post-metadata">

**Author:** ![rain](https://forum.dynare.org/letter_avatar_proxy/v4/letter/r/e95f7d/32.png) [@rain](https://forum.dynare.org/u/rain)\
**Post date:** [11 August 2010 18:48 UTC](https://forum.dynare.org/t/bayesian-irf/1594/3 "2010-08-11T18:48:44Z")

</div>

Thank you very much for the quick reply. It is clear to me now.

Best…