4

Suppose x is normally distributed with unknown mean and covariance matrix. Is there a set of conjugate priors for the mean and correlation matrix. I know that for covariance matrices one can use inverse Wishart distribution as priors, but what about a correlation matrix?

kjetil b halvorsen
  • 77,844
Mykie
  • 511
  • 2
    Knowing the correlation matrix is equal to $R$ doesn't fully specify the distribution, since there are any number of scalings consistent with that correlation. Is there a particular factorization of the covariance you have in mind, ie, $X \sim$Normal$(\mu, RD)$, for $D$ a diagonal matrix giving variances and $R$ positive definite matrix giving correlations? – Andrew M Nov 13 '14 at 08:33
  • 2
    If you let conjugacy go (because we have [more] efficient sampling algorithms these days), you can use the LKJ distribution, which is a distribution of correlation matrices. – Sycorax Nov 14 '14 at 00:29
  • I was about to name the LKJ prior as well... – Durden Jun 12 '23 at 20:35

0 Answers0