0
$\begingroup$

Suppose $\begin{bmatrix} K_{11} K_{12}\\K_{12}^T K_{22} \end{bmatrix}\sim\mathcal{IW}\left(\eta,\begin{bmatrix} \Sigma_{11} \Sigma_{12}\\\Sigma_{12}^T \Sigma_{22} \end{bmatrix}\right)$.

  1. What is the conditional distribution of $K_{11}|K_{22}$ and $K_{12}|K_{22}$
  2. What is the expectation of $K_{12}^TK_{22}^{-1}$
$\endgroup$

closed as unclear what you're asking by Stefan Kohl, Steven Sam, Alexandre Eremenko, John Pardon, S. Carnahan Jul 22 '14 at 12:59

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ the question "what is..." has an answer starting from the definition of the Inverse Wishart distribution, but I can't see how this would provide a simple closed-form expression, and I presume that's your real question. – $\endgroup$ – Carlo Beenakker Mar 13 '14 at 9:18
  • $\begingroup$ I was certainly looking for closed form solutions. Not sure what you mean by "what is..." $\endgroup$ – sachinruk Mar 14 '14 at 6:05
  • $\begingroup$ well, since you know the probability distribution of the full matrix, the marginal distribution of a set of matrix elements is obtained upon integration, and then the conditional distribution follows as well; so you know "what is the conditional distribution", it's just that you will not be able to carry out these integrations with paper and pencil to arrive at some tractable expression. $\endgroup$ – Carlo Beenakker Mar 14 '14 at 9:03
1
$\begingroup$

Have a look at the book Multivariate Statistics : A Vector Space Approach

by Morris L. Eaton, Chapter 8 pages 302-333, Open Access link.

It contains the results you need. Let me know if you need me to expand further.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.