# Deriving IRFs for different shocks

**URL:** <https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674>\
**Category:** Occbin\
**Created:** [3 December 2023 19:42 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674 "2023-12-03T19:42:07Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Qiao](https://forum.dynare.org/letter_avatar_proxy/v4/letter/q/ecccb3/32.png) [@Qiao](https://forum.dynare.org/u/Qiao)\
**Post date:** [3 December 2023 19:42 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674/1 "2023-12-03T19:42:07Z")

</div>

Hello everyone, I am new to Dynare and have a maybe dumb question here. I have already written the mod file and successfully generated some IRFs. But my goal is to draw graphs of IRFs of several different shocks (e.g. IRF of 1 unit of shock A, IRF of 1 unit of shock B,…) So far I can only do this by modifying the mod file everytime I want a new IRF, and I wonder if there are any easier ways (like using a loop) to do it. Thank you for your time!

---

<div class="post-metadata">

**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\
**Post date:** [4 December 2023 12:48 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674/2 "2023-12-04T12:48:01Z")

</div>

You should be getting an IRF for each shock declared in the `shocks`-block.

---

<div class="post-metadata">

**Author:** ![Qiao](https://forum.dynare.org/letter_avatar_proxy/v4/letter/q/ecccb3/32.png) [@Qiao](https://forum.dynare.org/u/Qiao)\
**Post date:** [4 December 2023 14:40 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674/3 "2023-12-04T14:40:51Z")

</div>

Thanks for your answers, professor, but when I include two shocks in the `shocks` -block, it ends up showing only one graph. Below is my code after laying out the model equations:

> steady;
> 
> check;
> 
> shocks(surprise);
> 
> var efc;
> 
> periods 1:4,5;
> 
> values 0,1;
> 
> var ete;
> 
> periods 1:4,5;
> 
> values 0,1;
> 
> end;
> 
> occbin\_setup;
> 
> occbin\_solver(simul\_periods=20,simul\_check\_ahead\_periods=200);
> 
> occbin\_graph logYe logYc logCe logCc logIe logIc CPIe CPIc;

---

<div class="post-metadata">

**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\
**Post date:** [4 December 2023 14:54 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674/4 "2023-12-04T14:54:50Z")

</div>

Your question was posted in the wrong category. For OccBin that is indeed not feasible. You need to conduct a separate simulation for each shock scenario as there is no recursive policy function to use.

---

<div class="post-metadata">

**Author:** ![Qiao](https://forum.dynare.org/letter_avatar_proxy/v4/letter/q/ecccb3/32.png) [@Qiao](https://forum.dynare.org/u/Qiao)\
**Post date:** [4 December 2023 15:43 UTC](https://forum.dynare.org/t/deriving-irfs-for-different-shocks/24674/5 "2023-12-04T15:43:58Z")

</div>

Thank you professor, and sorry for my wrong categorizing.