Learn R Programming

EESPCA (version 0.7.0)

Eigenvectors from Eigenvalues Sparse Principal Component Analysis (EESPCA)

Description

Contains logic for computing sparse principal components via the EESPCA method, which is based on an approximation of the eigenvector/eigenvalue identity. Includes logic to support execution of the TPower and rifle sparse PCA methods, as well as logic to estimate the sparsity parameters used by EESPCA, TPower and rifle via cross-validation to minimize the out-of-sample reconstruction error. H. Robert Frost (2021) .

Copy Link

Version

Install

install.packages('EESPCA')

Monthly Downloads

235

Version

0.7.0

License

GPL (>= 2)

Maintainer

H Robert Frost

Last Published

June 15th, 2022

Functions in EESPCA (0.7.0)

computeResidualMatrix

Calculates the residual matrix from the reduced rank reconstruction
computeApproxNormSquaredEigenvector

Approximates the normed squared eigenvector loadings
tpower

Implementation of the Yuan and Zhang TPower method.
riflePCACV

Sparsity parameter selection via cross-validation for rifle method of Tan et al.
EESPCA-package

Eigenvectors
eespca

Eigenvectors from Eigenvalues Sparse Principal Component Analysis (EESPCA)
powerIteration

Power iteration method for calculating principal eigenvector and eigenvalue.
reconstruct

Calculates the reduced rank reconstruction
eespcaForK

Multi-PC version of Eigenvectors from Eigenvalues Sparse Principal Component Analysis (EESPCA)
eespcaCV

Cross-validation for Eigenvectors from Eigenvalues Sparse Principal Component Analysis (EESPCA)
reconstructionError

Calculates the reduced rank reconstruction error
tpowerPCACV

Sparsity parameter selection for the Yuan and Zhang TPower method using cross-validation.
rifleInit

Computes the initial eigenvector for the rifle method of Tan et al.