# Get NaN result after Dynare simulation

**URL:** <https://forum.dynare.org/t/get-nan-result-after-dynare-simulation/18904>\
**Category:** Stochastic simulations\
**Created:** [17 September 2021 03:54 UTC](https://forum.dynare.org/t/get-nan-result-after-dynare-simulation/18904 "2021-09-17T03:54:14Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![xin\_sun](https://forum.dynare.org/letter_avatar_proxy/v4/letter/x/91b2a8/32.png) [@xin\_sun](https://forum.dynare.org/u/xin_sun)\
**Post date:** [17 September 2021 03:54 UTC](https://forum.dynare.org/t/get-nan-result-after-dynare-simulation/18904/1 "2021-09-17T03:54:14Z")

</div>

Hi there,

I am new to the Dynare, I want to drop the first 100 observations and use the next 200 observations of the simulation data. However, I don’t know why I get 0 in the matrix of covariance of endogenous shocks, therefore, the other standard deviations are all NaN, there is no reported error by dynare. I know that means they are all constant, but I don’t know how to modify the model to get the standard deviation.[nk\_simul\_MLE.mod](https://forum.dynare.org/uploads/short-url/wWSIAnPX5YmwVvWkhYel7Olmj7m.mod) (1.6 KB)

Any help will be appreciated. Thanks in advance.  
Warm regards,  
Xin

---

<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:** [17 September 2021 09:39 UTC](https://forum.dynare.org/t/get-nan-result-after-dynare-simulation/18904/2 "2021-09-17T09:39:34Z")

</div>

You don’t have a `shocks`-block to set the standard deviations. It is required now, because you don’t estimate the standard deviations.

---

<div class="post-metadata">

**Author:** ![xin\_sun](https://forum.dynare.org/letter_avatar_proxy/v4/letter/x/91b2a8/32.png) [@xin\_sun](https://forum.dynare.org/u/xin_sun)\
**Post date:** [17 September 2021 12:08 UTC](https://forum.dynare.org/t/get-nan-result-after-dynare-simulation/18904/3 "2021-09-17T12:08:27Z")

</div>

Thanks for your quick reply, I see the point now, thanks a lot.