# Ordinal regression in brms: category-specific effects for cumulative family

**URL:** https://discourse.mc-stan.org/t/ordinal-regression-in-brms-category-specific-effects-for-cumulative-family/7646
**Category:** brms
**Created:** [February 9, 2019, 1:39pm UTC](https://discourse.mc-stan.org/t/ordinal-regression-in-brms-category-specific-effects-for-cumulative-family/7646 "2019-02-09T13:39:25Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![Homer](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/homer/32/4172_2.png) [@Homer](https://discourse.mc-stan.org/u/Homer)
#### Post date: [February 9, 2019, 1:39pm UTC](https://discourse.mc-stan.org/t/ordinal-regression-in-brms-category-specific-effects-for-cumulative-family/7646/1 "2019-02-09T13:39:25Z")

</div>

For the cumulative family, brms doesn’t support category-specific effects. However, in the Stan code that can be saved when using brm() with “family = cumulative()”, I found the functions cumulative\_logit\_lpmf() and cumulative\_logit\_cs\_lpmf(), the second one giving a hint that category-specific effects could be possible for the cumulative family when replacing the call to ordered\_logistic\_lpmf() in the model block by a call to cumulative\_logit\_cs\_lpmf(). So my question is: Why are category-specific effects not supported for the cumulative family? I guess there is a proper reason but I can’t see it yet.

---

<div class="post-metadata">

### Author: ![paul.buerkner](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/paul.buerkner/32/3303_2.png) [@paul.buerkner](https://discourse.mc-stan.org/u/paul.buerkner)
#### Post date: [February 9, 2019, 1:44pm UTC](https://discourse.mc-stan.org/t/ordinal-regression-in-brms-category-specific-effects-for-cumulative-family/7646/2 "2019-02-09T13:44:49Z")

</div>

cumulative models with category specific effects can imply negative probabilities.