# Simulation with confidence intervals

**URL:** <https://forum.certara.com/t/simulation-with-confidence-intervals/2344>\
**Category:** Modelling and Simulation\
**Created:** [December 8, 2011, 3:44pm UTC](https://forum.certara.com/t/simulation-with-confidence-intervals/2344 "2011-12-08T15:44:35Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![charliebrindley](https://avatars.discourse-cdn.com/v4/letter/c/f14d63/32.png) [@charliebrindley](https://forum.certara.com/u/charliebrindley)\
**Post date:** [December 8, 2011, 3:44pm UTC](https://forum.certara.com/t/simulation-with-confidence-intervals/2344/1 "2011-12-08T15:44:35Z")

</div>

I have fitted a PK model using classic WNL to pooled plasma concentrations (naive-pooled) from 6 subjects each receiving 7 daily doses (data available on Days 1 and 7). How can I simulate a profile (e.g. for 14 daily doses) from the derived parameters with 95% CIs? Do I have to take the CI for each model parameter and simulate a lower and upper CI profile separately?

---

<div class="post-metadata">

**Author:** ![blakester](https://avatars.discourse-cdn.com/v4/letter/b/57b2e6/32.png) [@blakester](https://forum.certara.com/u/blakester)\
**Post date:** [May 1, 2012, 3:08pm UTC](https://forum.certara.com/t/simulation-with-confidence-intervals/2344/2 "2012-05-01T15:08:10Z")

</div>

I am working on a similar problem at the moment. I have modelled individual data. My plan is to get the 95% CI of the derived parameters across individuals then take upper and lower bounds for simulation purposes. Did you achieve a solution for your original post?

---

<div class="post-metadata">

**Author:** ![charliebrindley](https://avatars.discourse-cdn.com/v4/letter/c/f14d63/32.png) [@charliebrindley](https://forum.certara.com/u/charliebrindley)\
**Post date:** [November 20, 2012, 12:41pm UTC](https://forum.certara.com/t/simulation-with-confidence-intervals/2344/3 "2012-11-20T12:41:07Z")

</div>

Sorry to reply so late! I couldn’t work out how to do this, could you?

---

<div class="post-metadata">

**Author:** ![blakester](https://avatars.discourse-cdn.com/v4/letter/b/57b2e6/32.png) [@blakester](https://forum.certara.com/u/blakester)\
**Post date:** [November 27, 2012, 3:46pm UTC](https://forum.certara.com/t/simulation-with-confidence-intervals/2344/4 "2012-11-27T15:46:24Z")

</div>

No, used a very empirical approach. Used individually predicted PK parameters from each subject in simulation. From the resultant predicted plasma concentration-time profiles estimated 5th and 95th percentile.