Application of multiple comparison type procedures to the eigenvalues of §-11:§2:
The purpose of this paper is to investigate the possible application of multiple comparisons type procedures for inferring differences among the eigenvalues of \(\Sigma_ 1^{-1}\Sigma_ 2\) where \(\Sigma_ 1\) and \(\Sigma_ 2\) are the covariance matrices for two multivariate normal populations. The decision rules used in the inference procedures are based upon the eigenvalues of \(S_ 1^{-1}S_ 2\) where \(S_ 1\) and \(S_ 2\) are the maximum likelihood estimates for \(\Sigma_ 1\) and \(\Sigma_ 2\), respectively. Certain theoretical results are presented giving justification to the multiple comparisons procedures presented. Finally, due to the complex nature of the distribution of the eigenvalues of \(S_ 1^{-1}S_ 2\) the properties of the procedures are exhibited by way of simulation analysis on the trivariate problem.
- Estimation of the eigenvalues of \(\Sigma{}_ 1\Sigma{}_ 2^{-1}\)
- The comparison of sample covariance matrices using likelihood ratio tests
- A STATISTIC FOR TESTING THE EQUALITY OF EIGENVALUE OF COVARIANCE MATRIX ON MULTIPOPULATION
- Magnitudinal effects in the normal multivariate model
- Simultaneous procedures for covariance matrices
- Combinatoric classification of multivariate normal variates
- Distributions of Matrix Variates and Latent Roots Derived from Normal Samples
- Simultaneous tests for equality of latent roots against certain alternatives. I
- Some Distribution Problems Connected with the Characteristic Roots of $S_1S^{-1}_2$
- Some relations between the comparison of covariance matrices and principal component analysis
This page was built for publication: Application of multiple comparison type procedures to the eigenvalues of §-11:§2:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3747532)