Large sample approximations for the LR statistic for equality of the smallest eigenvalues of a covariance matrix under elliptical population.
Daisuke WatanabeSusumu OkadaYasunori FujikoshiTakakazu SugiyamaPublished in: Comput. Stat. Data Anal. (2008)
Keyphrases
- covariance matrix
- sample size
- covariance matrices
- eigenvalues and eigenvectors
- correlation matrix
- principal component analysis
- normal distribution
- gaussian mixture
- mahalanobis distance
- closed form
- class conditional densities
- multivariate gaussian
- eigendecomposition
- positive definite
- model selection
- pseudo inverse
- objective function
- worst case
- support vector
- symmetric matrix
- geometrical interpretation
- machine learning