bdsvd: Block Structure Detection Using Singular Vectors

Provides methods to perform block diagonal covariance matrix detection using singular vectors ('BD-SVD'), which can be extended to inherently sparse principal component analysis ('IS-PCA'). The methods are described in Bauer (2025) <doi:10.1080/10618600.2024.2422985> and Bauer (2026) <doi:10.48550/arXiv.2510.03729>.

Version: 1.2.1
Imports: irlba, matrixStats, Rcpp, stats
LinkingTo: Rcpp, RcppArmadillo
Suggests: cvCovEst, glasso, mvtnorm, dslabs, testthat (≥ 3.0.0)
Published: 2026-03-26
DOI: 10.32614/CRAN.package.bdsvd
Author: Jan O. Bauer ORCID iD [aut, cre], Ron Holzapfel [aut]
Maintainer: Jan O. Bauer <j.bauer at vu.nl>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Materials: README
CRAN checks: bdsvd results

Documentation:

Reference manual: bdsvd.html , bdsvd.pdf

Downloads:

Package source: bdsvd_1.2.1.tar.gz
Windows binaries: r-devel: bdsvd_0.2.1.zip, r-release: bdsvd_0.2.1.zip, r-oldrel: bdsvd_0.2.1.zip
macOS binaries: r-release (arm64): bdsvd_0.2.1.tgz, r-oldrel (arm64): bdsvd_0.2.1.tgz, r-release (x86_64): bdsvd_0.2.1.tgz, r-oldrel (x86_64): bdsvd_0.2.1.tgz
Old sources: bdsvd archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=bdsvd to link to this page.