Improved method for correcting sample Mahalanobis distance without estimating population eigenvalues or eigenvectors of covariance matrix.
Yasuyuki KobayashiPublished in: Int. J. Data Sci. Anal. (2020)
Keyphrases
- covariance matrix
- mahalanobis distance
- covariance matrices
- sample size
- eigenvalues and eigenvectors
- correlation matrix
- eigendecomposition
- principal component analysis
- gaussian mixture
- relative entropy
- pseudo inverse
- principal components
- laplacian matrix
- objective function
- computer vision
- maximum likelihood
- computational complexity