Some tests for the covariance matrix with fewer observations than the dimension under non-normality.
Muni S. SrivastavaTõnu KolloDietrich von RosenPublished in: J. Multivar. Anal. (2011)
Keyphrases
- covariance matrix
- normal distribution
- covariance matrices
- sample size
- principal component analysis
- positive definite
- gaussian mixture
- geometrical interpretation
- multivariate gaussian
- objective function
- correlation matrix
- mahalanobis distance
- eigendecomposition
- pseudo inverse
- multivariate normal
- multi objective
- eigenvalues and eigenvectors
- estimation error
- feature selection