Improved Stein-type shrinkage estimators for the high-dimensional multivariate normal covariance matrix.
Thomas J. FisherXiaoqian SunPublished in: Comput. Stat. Data Anal. (2011)
Keyphrases
- covariance matrix
- multivariate normal
- high dimensional
- covariance matrices
- principal component analysis
- multivariate gaussian
- mahalanobis distance
- sample size
- low dimensional
- gaussian mixture
- positive definite
- objective function
- correlation matrix
- similarity search
- eigenvalues and eigenvectors
- principal components
- class conditional densities
- eigendecomposition
- dimensionality reduction
- distance measure
- feature selection
- vector space
- cma es
- parameter space
- maximum likelihood
- denoising
- least squares
- bayesian networks
- image processing