So basically, I’m interested in obtaining the sampling variance of the covariance of two random effects. Is there a way to obtain such estimate in brms?

Hi, it seems your question fell through a bit. However as posed, it is hard to answer - could you be more specific on what is your intended use case? My best guess is that for a model like `y ~ (a + b | g)`

you would take the posterior samples of the correlation matrix, multiply by the relevant standard deviations to get the covariance matrix and then compute the MCSE for the mean of this covariance - but I am not sure this matches what you need.