factorselect: Eigenvalue-Based Estimation of the Number of Factors in Approximate Factor Models

Eigenvalue-based estimation of the number of factors in approximate factor models. Designed to work when either N or T is large, without requiring both dimensions to grow simultaneously. Implements the eigenvalue ratio estimator of Ahn and Horenstein (2013) <doi:10.3982/ECTA8968>, the information criteria of Bai and Ng (2002) <doi:10.1111/1468-0262.00273>, the tuned penalty of Alessi, Barigozzi and Capasso (2010) <doi:10.1016/j.spl.2010.08.005>, the auto-covariance ratio estimator of Lam and Yao (2012) <doi:10.1214/12-AOS970>, and the edge distribution estimators of Onatski (2009) <doi:10.3982/ECTA6964> and Onatski (2010) <doi:10.1162/REST_a_00043>.

Version: 0.1.3
Suggests: RSpectra, testthat (≥ 3.0.0), knitr, rmarkdown
Published: 2026-04-28
DOI: 10.32614/CRAN.package.factorselect
Author: Jason Parker ORCID iD [aut, cre]
Maintainer: Jason Parker <jparker588 at gmail.com>
BugReports: https://github.com/penny4nonsense/factorselect/issues
License: MIT + file LICENSE
URL: https://github.com/penny4nonsense/factorselect
NeedsCompilation: no
Materials: README, NEWS
CRAN checks: factorselect results

Documentation:

Reference manual: factorselect.html , factorselect.pdf
Vignettes: Introduction to factorselect (source, R code)

Downloads:

Package source: factorselect_0.1.3.tar.gz
Windows binaries: r-devel: not available, r-release: factorselect_0.1.3.zip, r-oldrel: factorselect_0.1.3.zip
macOS binaries: r-release (arm64): factorselect_0.1.3.tgz, r-oldrel (arm64): factorselect_0.1.3.tgz, r-release (x86_64): not available, r-oldrel (x86_64): not available

Linking:

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