# Reparameterizing to avoid low E-BFMI warning

**URL:** <https://discourse.mc-stan.org/t/reparameterizing-to-avoid-low-e-bfmi-warning/7957>\
**Category:** Modeling\
**Tags:** fitting-issues\
**Created:** [March 6, 2019, 11:23pm UTC](https://discourse.mc-stan.org/t/reparameterizing-to-avoid-low-e-bfmi-warning/7957 "2019-03-06T23:23:20Z")\
**Posts on this page:** 1\
**Showing post:** 4

<div class="post-metadata">

**Author:** ![bbbales2](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/bbbales2/32/77_2.png) [@bbbales2](https://discourse.mc-stan.org/u/bbbales2)\
**Post date:** [March 8, 2019, 4:44am UTC](https://discourse.mc-stan.org/t/reparameterizing-to-avoid-low-e-bfmi-warning/7957/4 "2019-03-08T04:44:18Z")

</div>

> [@mdonoghoe](#):
>
> Thinking about it more, I agreed that these put too much weight on quite extreme values in my situation, so I changed it to a half-normal prior.

Nice! Glad to hear it’s working out.

> [@mdonoghoe](#):
>
> ‘design matrix’ of 0s and 1s

Aaah, gotcha, makes sense.

> [@mdonoghoe](#):
>
> that correlation matrix Ω\Omega is of great interest.

Someone in another thread just asked something about why you’d want to model a covariance like this. I think I messed up the answer. If you’re feeling charitable: [Estimating covariance terms in multilevel model - #2 by bbbales2](https://discourse.mc-stan.org/t/estimating-covariance-terms-in-multilevel-model/7940/2) :D

---

_[View the full topic](https://discourse.mc-stan.org/t/reparameterizing-to-avoid-low-e-bfmi-warning/7957)._