# Simulations using Sim./pred Check

**URL:** <https://forum.certara.com/t/simulations-using-sim-pred-check/3077>\
**Category:** Modelling and Simulation\
**Created:** [November 18, 2016, 9:50pm UTC](https://forum.certara.com/t/simulations-using-sim-pred-check/3077 "2016-11-18T21:50:50Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![vaishnavi.ganti](https://avatars.discourse-cdn.com/v4/letter/v/e68b1a/32.png) [@vaishnavi.ganti](https://forum.certara.com/u/vaishnavi.ganti)\
**Post date:** [November 18, 2016, 9:50pm UTC](https://forum.certara.com/t/simulations-using-sim-pred-check/3077/1 "2016-11-18T21:50:50Z")

</div>

Dear All,

I have a query with regards to using Simulation/Pred Check for the Pop PK model.

For simplicity lets assume its a naive pooled approach. Based on the structural model I now want to simulate for :

1. different dosing schemes

2. Different parameter estimates.

3. Also I would like to see the “uncertainty” of these simulations ( a CI band around the simulations)

Any help would be appreciated

Thank you so much

---

<div class="post-metadata">

**Author:** ![minhaj](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@minhaj](https://forum.certara.com/u/minhaj)\
**Post date:** [November 19, 2016, 1:42pm UTC](https://forum.certara.com/t/simulations-using-sim-pred-check/3077/2 "2016-11-19T13:42:29Z")

</div>

the first way to add uncertainty in the average is to use the theta table with cv in %. divide by 100 and take the square. This will be an estimate of the variance in the log-domaine but without taing into account the correlation between V and Cl. With this method, you need to sue the default log-normal for parameters  
If you use the table called thetacovariance, it has the variance and covariance uncertainty in terms of a variance covariance. Here you need to use the normal assumption for your parameters as this var covar is not in the log-domain.  
You can create any simulation template with the dosing regimen you want and then use your best estimate you got from the original fit and use either one of the 2 tecniques I suggested above to simulate the uncertainty about your average profile. I am trying to send you the project but it does not work

---

<div class="post-metadata">

**Author:** ![minhaj](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@minhaj](https://forum.certara.com/u/minhaj)\
**Post date:** [November 19, 2016, 5:07pm UTC](https://forum.certara.com/t/simulations-using-sim-pred-check/3077/3 "2016-11-19T17:07:21Z")

</div>

Here we go

[sim\_average\_with\_uncertainty\_method\_theta\_cv.phxproj](https://forum.certara.com/uploads/short-url/bD7smgKRPngFzLhUUPffXMwmapC.phxproj) (5.55 MB)[sim\_average\_with\_uncertainty\_method\_thetacovariance.phxproj](https://forum.certara.com/uploads/short-url/aU8o4AInyT2rppHeiGRpVBaHARF.phxproj) (5.55 MB)