Say I have a Matrix-Normal distribution and two Inverse Wishart Distributions

$$X \sim MN_{p\times n}(0, \Sigma, \Omega)$$ $$ \Sigma \sim IW(a, A) $$ $$ \Omega \sim IW(b, B)$$ where $a$ and $b$ are degrees of freedom and $A$ and $B$ are scale matrices.

I know that the marginal distribution of $p(X, \Omega)$ is matrix-T and given by $ T_{p\times n} (a, 0, A, \Omega)$. What is the marginal distribution of $p(X)$?

In other words, what is the marginal distribution when you marginalize over both inverse Wishart random variables? Is there a name for this distribution?

  • $\begingroup$ Did this end up having a nice posterior distribution? That is, is this conjugate? $\endgroup$ – shabbychef Feb 13 at 6:34
  • $\begingroup$ I think this is not going to have a nice form because the Kronecker product of Inverse Wisharts is not a known distribution. $\endgroup$ – shabbychef Feb 13 at 6:35
  • $\begingroup$ Agreed. I doubt its nice, but this remains an outstanding question. I don't have an answer yet. $\endgroup$ – jds Feb 17 at 13:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.