Missing IRF for a shock

Hi,

When trying to use the stoch_sim function in this model: Groll_Monacelli_2018_HJ.mod (4.3 KB), I only get the IRF of “pih” and “pistar” for the markup shocks (i.e. “emu” and “emustar”), but not for the shocks to the natural level of output (i.e. “z” and “zstar”). I thought that this might be caused by some “divine coincidence”, meaning that the Dynare won’t produce the IRFs due to too little respons in “pih” and “pistar”. Is that the reason? or do you know what else could explain this?

I hope that you can help.
Thanks!

Have a look at the decision rules. They clearly show that those two shocks do not have an impact on those variables. Whether this is desired or a bug in your code is something you need to judge.