# How can I solve "bfmi-low" problem?

**URL:** <https://discourse.mc-stan.org/t/how-can-i-solve-bfmi-low-problem/3018>\
**Category:** Modeling\
**Created:** [January 17, 2018, 2:21pm UTC](https://discourse.mc-stan.org/t/how-can-i-solve-bfmi-low-problem/3018 "2018-01-17T14:21:56Z")\
**Posts on this page:** 1\
**Showing post:** 14

<div class="post-metadata">

**Author:** ![avehtari](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/avehtari/32/5935_2.png) [@avehtari](https://discourse.mc-stan.org/u/avehtari)\
**Post date:** [January 17, 2018, 6:15pm UTC](https://discourse.mc-stan.org/t/how-can-i-solve-bfmi-low-problem/3018/14 "2018-01-17T18:15:14Z")

</div>

> [@ignacio](#):
>
> I tried normal(0,10) and normal(0,1), but the problem does not go away.

Those are really wide compared to the original posterior. Right now I’m not worried that you would use too informative prior, now I just want to know what happens if you have informative prior. If that helps with the computation then you can start thinking harder about the priors or accept that sampling is going to be slow. Also you should think anyway whether that another smaller mode is plausible.

---

_[View the full topic](https://discourse.mc-stan.org/t/how-can-i-solve-bfmi-low-problem/3018)._