# LKJCholeskyCov shape argument

**URL:** https://discourse.pymc.io/t/lkjcholeskycov-shape-argument/7270
**Category:** Questions
**Created:** [April 18, 2021, 3:02am UTC](https://discourse.pymc.io/t/lkjcholeskycov-shape-argument/7270 "2021-04-18T03:02:45Z")
**Posts on this page:** 1
**Showing post:** 10

<div class="post-metadata">

### Author: ![jonsedar](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/jonsedar/32/2590_2.png) [@jonsedar](https://discourse.pymc.io/u/jonsedar)
#### Post date: [April 20, 2021, 10:40am UTC](https://discourse.pymc.io/t/lkjcholeskycov-shape-argument/7270/10 "2021-04-20T10:40:47Z")

</div>

A quick crossref to the ever knowledgeable @junpenglao would suggest that yes: it’s common practice to loop the LKJ priors [LKJCholeskyCov input dependent](https://discourse.pymc.io/t/lkjcholeskycov-input-dependent/1570)

I hear you on the not quite elegant bit, and to jump on from Junpeng’s note, with your participants x trials dataset, I hazard that you might be able to treat this as a [pm.MatrixNormal](https://docs.pymc.io/api/distributions/multivariate.html#pymc3.distributions.multivariate.MatrixNormal), though TBH I’m not familiar with them in practice. I found a good explanation here: [Matrix Variate Normal Distributions with MixMatrix](https://cran.r-project.org/web/packages/MixMatrix/vignettes/matrixnormal.html)

Re `uij`, if you’re setting a different `chol` per participant (is this `i` ?), then wouldn’t;t you have a `uj`? FWIW you can probably index 2D if you want to `u[i][j]`, but that’ll probably be a level of complication you dont need

---

_[View the full topic](https://discourse.pymc.io/t/lkjcholeskycov-shape-argument/7270)._