# Understanding LOOIC

**URL:** <https://discourse.mc-stan.org/t/understanding-looic/13409>\
**Category:** Modeling\
**Tags:** loo, interpret-results, cognitive-science\
**Created:** [February 29, 2020, 12:51pm UTC](https://discourse.mc-stan.org/t/understanding-looic/13409 "2020-02-29T12:51:21Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![andre.pfeuffer](https://avatars.discourse-cdn.com/v4/letter/a/e480ec/32.png) [@andre.pfeuffer](https://discourse.mc-stan.org/u/andre.pfeuffer)\
**Post date:** [February 29, 2020, 1:20pm UTC](https://discourse.mc-stan.org/t/understanding-looic/13409/2 "2020-02-29T13:20:43Z")

</div>

For `looic`, lower numbers are better. What you want is to compare the looic values of other models. having same data, ~~likelihood~~ (and seed). If you look at your comparison, `elpd_diff` equals -1.9, that is the difference between the two models and the difference se\_diff is 0.3 is significant, because:

0 \notin[- 1.9 - 0.3 , -1.9 + 0.3]  
looic = -2 \* elpd\_{loo}

`p_loo` is explained in detail here:

> [@A quick note what I infer from p\_loo and Pareto k values](https://discourse.mc-stan.org/t/a-quick-note-what-i-infer-from-p-loo-and-pareto-k-values/3446):
>
> This a quick note on different cases why Pareto k values can be large. I’ll try to extend this to a longer explanation and a case study. Any feedback on this draft is appreciated If all Pareto k small, model is likely to be ok (although there can be better models) If high Pareto k values If p\_loo \<\< the number of parameters p, then the model is likely to be misspecified. PPC is likely to detect the problem, too. Try using overdispersed model, or add more structural information (nonlinearity, …

---

_[View the full topic](https://discourse.mc-stan.org/t/understanding-looic/13409)._