# Translating robust logistic regression from rstan to brms

**URL:** <https://discourse.mc-stan.org/t/translating-robust-logistic-regression-from-rstan-to-brms/5244>\
**Category:** brms\
**Tags:** rstan\
**Created:** [August 16, 2018, 4:47pm UTC](https://discourse.mc-stan.org/t/translating-robust-logistic-regression-from-rstan-to-brms/5244 "2018-08-16T16:47:58Z")\
**Posts on this page:** 1\
**Showing post:** 8

<div class="post-metadata">

**Author:** ![andrewgelman](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/andrewgelman/32/1577_2.png) [@andrewgelman](https://discourse.mc-stan.org/u/andrewgelman)\
**Post date:** [March 10, 2019, 10:33pm UTC](https://discourse.mc-stan.org/t/translating-robust-logistic-regression-from-rstan-to-brms/5244/8 "2019-03-10T22:33:36Z")

</div>

Hi, yes, I guess this would have a problem if nu is less than 2, so you’d want to constrain nu to be at least 2, or even to constrain it to be at least 3. I don’t know how to ping Aki on this so I’ll just cc him, also Jonah and Ben in case this ever came up in rstanarm.  
See you

Andrew

---

_[View the full topic](https://discourse.mc-stan.org/t/translating-robust-logistic-regression-from-rstan-to-brms/5244)._