Approximate null distribution of the largest root in multivariate analysis
Open Access
- 1 December 2009
- journal article
- Published by Institute of Mathematical Statistics in The Annals of Applied Statistics
- Vol. 3 (4) , 1616-1633
- https://doi.org/10.1214/08-aoas220
Abstract
The greatest root distribution occurs everywhere in classical multivariate analysis, but even under the null hypothesis the exact distribution has required extensive tables or special purpose software. We describe a simple approximation, based on the Tracy–Widom distribution, that in many cases can be used instead of tables or software, at least for initial screening. The quality of approximation is studied, and its use illustrated in a variety of settings.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Multivariate analysis and Jacobi ensembles: Largest eigenvalue, Tracy–Widom limits and rates of convergenceThe Annals of Statistics, 2008
- Universality results for the largest eigenvalues of some sample covariance matrix ensemblesProbability Theory and Related Fields, 2008
- Roy Table: A Program for Generating Tables of Critical Values for Roy’s Largest Root CriterionEducational and Psychological Measurement, 2000
- On orthogonal and symplectic matrix ensemblesCommunications in Mathematical Physics, 1996
- Upper percentage points of the largest root of a matrix in multivariate analysisBiometrika, 1967
- On the distribution of the largest of six roots of a matrix in multivariate analysisBiometrika, 1959
- UPPER PERCENTAGE POINTS OF THE GENERALIZED BETA DISTRIBUTION. IIIBiometrika, 1958
- Some Results Useful in Multivariate AnalysisThe Annals of Mathematical Statistics, 1956
- ON THE DISTRIBUTION OF THE LARGEST OR THE SMALLEST ROOT OF A MATRIX IN MULTIVARIATE ANALYSISBiometrika, 1956
- Regressions between Sets of VariablesEconometrica, 1942