# Multivariate normal data covariance estimation

**URL:** <https://discourse.mc-stan.org/t/multivariate-normal-data-covariance-estimation/14995>\
**Category:** Modeling\
**Tags:** multivariate-normal, performance, covariance\
**Created:** [May 9, 2020, 6:24pm UTC](https://discourse.mc-stan.org/t/multivariate-normal-data-covariance-estimation/14995 "2020-05-09T18:24:56Z")\
**Posts on this page:** 1\
**Showing post:** 4

<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:** [May 10, 2020, 12:31pm UTC](https://discourse.mc-stan.org/t/multivariate-normal-data-covariance-estimation/14995/4 "2020-05-10T12:31:51Z")

</div>

[https://discourse.mc-stan.org/t/centered-or-non-centered-parametrization-for-random-effects/728](https://discourse.mc-stan.org/t/centered-or-non-centered-parametrization-for-random-effects/728)

Right. See Bob’s answers in above link.  
“… for `y` being data; the built in `y ~ normal(mu, sigma)` will be better”

Thanks for closing the topic.

---

_[View the full topic](https://discourse.mc-stan.org/t/multivariate-normal-data-covariance-estimation/14995)._