No eigenvalues outside the support of the limiting empirical spectral distribution of a separable covariance matrix.
Debashis PaulJack W. SilversteinPublished in: J. Multivar. Anal. (2009)
Keyphrases
- covariance matrix
- normal distribution
- covariance matrices
- principal component analysis
- eigenvalues and eigenvectors
- laplacian matrix
- positive definite
- sample size
- pseudo inverse
- mahalanobis distance
- correlation matrix
- gaussian mixture
- objective function
- eigendecomposition
- geometrical interpretation
- principal components
- probability distribution
- multivariate gaussian
- mahalanobis metric
- gaussian distribution
- spectral analysis
- gaussian mixture model
- distance measure
- dimensionality reduction
- computational complexity
- image processing
- symmetric matrix