I am interested in finding expressions for the marginal distributions of the off-diagonal terms in a Wishart-distributed random variable.

More specifically, suppose $X$ is an $n \times p$ matrix, each row of which is independently drawn from a $p$-variate normal distribution with zero mean: $X_{(i)}{=}(x_i^1,\dots,x_i^p)^T\sim N_p(0,V)$. Then $S=X^T X$ is a Wishart-distributed random variable, denoted $S\sim W_p(V,n)$.

Expressions exist for the marginal distributions of the diagonal terms, $S_{i,i}$ (e.g., page 4, third bullet point in this document). But I haven't been able to find similar expressions for the off-diagonal marginal distributions, $S_{i,j}$, where $i \neq j$.

Do they exist and, if so, what are they?

  • $\begingroup$ I wonder if you can rotate the problem away. That is, consider the distribution of $QSQ^{\top}$ for orthonormal $Q$, where you have somehow rotated an off diagonal onto a diagonal? $\endgroup$
    – shabbychef
    Sep 18, 2014 at 16:19

1 Answer 1


I believe the answer to this can actually be found on the wikipedia article about the Wishart distribution, which indicates that it is variance-gamma distributed.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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