# Strategies for graphing distributions of log-odds estimates and the corresponding odds ratios

**URL:** <https://discourse.mc-stan.org/t/strategies-for-graphing-distributions-of-log-odds-estimates-and-the-corresponding-odds-ratios/11981>\
**Category:** Modeling\
**Tags:** techniques\
**Created:** [November 19, 2019, 4:48am UTC](https://discourse.mc-stan.org/t/strategies-for-graphing-distributions-of-log-odds-estimates-and-the-corresponding-odds-ratios/11981 "2019-11-19T04:48:54Z")\
**Posts on this page:** 1\
**Showing post:** 9

<div class="post-metadata">

**Author:** ![mike-lawrence](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/mike-lawrence/32/59_2.png) [@mike-lawrence](https://discourse.mc-stan.org/u/mike-lawrence)\
**Post date:** [December 5, 2019, 2:06pm UTC](https://discourse.mc-stan.org/t/strategies-for-graphing-distributions-of-log-odds-estimates-and-the-corresponding-odds-ratios/11981/9 "2019-12-05T14:06:10Z")

</div>

> [@llewmills](#):
>
> Do you think a similar strategy for the HPDI, i.e. get the upper and lower bounds of the HPDI for the distribution of log odds and then exponentiate those values? Or do you think it’s safe to exponentiate the difference in estimated log-odds at each step in the mcmc chain and then take the HPDI of the resultant chain of odds ratios?

Note that there are [good arguments against HPDIs](https://discourse.mc-stan.org/t/quantiles-vs-hpdi-from-rstan-summary/9025). If you used straight quantiles, they’ll be transformation invariant.

---

_[View the full topic](https://discourse.mc-stan.org/t/strategies-for-graphing-distributions-of-log-odds-estimates-and-the-corresponding-odds-ratios/11981)._