Hello, I’m writing “toy” Gibbs sampler and my idea was to use LKJCholeskyCov for covariance matrix prior in “stand alone mode” (without pymc model). But seems it has no “random” method like others distributions. So my question is why and is it possible to use LKJ distribution outside of a model. T…

yep LKJCholeskyCov does not have a random method. I think you can try modifying the LKJCorr random method : generate a random correlation matrix using LKJCorr random method, and transform the sample, then combine with the random generated diagonal matrix using sd_dist.

LKJCholeskyCov random method

Questions

Kyrish August 28, 2018, 7:34pm 3

Thanks! That’s the way to go.

Topic		Replies	Views
Uses of LKJCholeskyCov and LKJCorr Questions	2	2286	July 26, 2019
Covariate matrices as observations Questions	6	870	October 27, 2020
Which variables show up in sampling chain and which don't? Questions	5	626	July 24, 2018
LKJCholeskyCov input dependent Questions	1	564	July 24, 2018
Call MvNormal.random() method inside model Questions	3	1159	August 29, 2018

LKJCholeskyCov random method

Related topics