## Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

## If You Use a Screen Reader

This content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.

# Stochastic Majorization of the Log-Eigenvalues of a Bivariate Wishart Matrix

Michael D. Perlman
Lecture Notes-Monograph Series
Vol. 5, Inequalities in Statistics and Probability (1984), pp. 173-177
Stable URL: http://www.jstor.org/stable/4355497
Page Count: 5
Let $l=(l_{1},l_{2})$ and $\lambda =(\lambda _{1},\lambda _{2})$, where $\lambda _{1}\geq \lambda _{2}>0$ are the ordered eigenvalues of S and Σ, respectively, and ${\bf S}\sim W_{2}(n,\Sigma)$ is a bivariate Wishart matrix. Let m = (m1,m2) and $\mu =(\mu _{1},\mu _{2})$, where $m_{i}=\text{log}\ l_{i}$ and $\mu _{i}=\text{log}\ \lambda _{i}$. It is shown that $P_{\mu}\{{\bf m}\not\in B\}$ is Schur-convex in μ whenever B is a Schur-monotone set, i.e. [x ∈ B, x majorizes ${\bf x}^{\ast}$] ⇒ ${\bf x}^{\ast}\in B$. This result implies the unbiasedness and power-monotonicity of a class of invariant tests for bivariate sphericity and other orthogonally invariant hypotheses.