Matrix variate beta distribution

Last updated
Matrix variate beta distribution
Notation
Parameters
Support matrices with both and positive definite
PDF
CDF

In statistics, the matrix variate beta distribution is a generalization of the beta distribution. It is also called the MANOVA ensemble and the Jacobi ensemble.

Contents

If is a positive definite matrix with a matrix variate beta distribution, and are real parameters, we write (sometimes ). The probability density function for is:

Here is the multivariate beta function:

where is the multivariate gamma function given by

Theorems

Distribution of matrix inverse

If then the density of is given by

provided that and .

Orthogonal transform

If and is a constant orthogonal matrix, then

Also, if is a random orthogonal matrix which is independent of , then , distributed independently of .

If is any constant , matrix of rank , then has a generalized matrix variate beta distribution, specifically .

Partitioned matrix results

If and we partition as

where is and is , then defining the Schur complement as gives the following results:

Wishart results

Mitra proves the following theorem which illustrates a useful property of the matrix variate beta distribution. Suppose are independent Wishart matrices . Assume that is positive definite and that . If

where , then has a matrix variate beta distribution . In particular, is independent of .

Spectral density

The spectral density is expressed by a Jacobi polynomial. [1]

Extreme value distribution

The distribution of the largest eigenvalue is well approximated by a transform of the Tracy–Widom distribution. [2]

See also

References

  1. ( Potters & Bouchaud 2020 )
  2. Johnstone, Iain M. (2008-12-01). "Multivariate analysis and Jacobi ensembles: Largest eigenvalue, Tracy–Widom limits and rates of convergence". The Annals of Statistics. 36 (6). doi:10.1214/08-AOS605. ISSN   0090-5364.