PAGFL: Joint Estimation and Identification of Latent Groups in Panel Data Models

In panel data analysis, unobservable group structures are a common challenge. Disregarding group-level heterogeneity by assuming an entirely homogeneous panel can introduce bias. Conversely, estimating individual coefficients for each cross-sectional unit is inefficient and may lead to high uncertainty. This package addresses this issue by implementing the pairwise adaptive group fused Lasso (PAGFL) by Mehrabani (2023) <doi:10.1016/j.jeconom.2022.12.002>. PAGFL is an efficient methodology to identify latent group structures and estimate group-specific coefficients simultaneously.

Version: 1.0.1
Imports: Rcpp, pbapply
LinkingTo: Rcpp, RcppArmadillo
Published: 2024-02-17
Author: Paul Haimerl ORCID iD [aut, cre], Ali Mehrabani ORCID iD [ctb]
Maintainer: Paul Haimerl <paul.haimerl at maastrichtuniversity.nl>
BugReports: https://github.com/Paul-Haimerl/PAGFL/issues
License: AGPL (≥ 3)
URL: https://github.com/Paul-Haimerl/PAGFL
NeedsCompilation: yes
Materials: README NEWS
CRAN checks: PAGFL results

Documentation:

Reference manual: PAGFL.pdf

Downloads:

Package source: PAGFL_1.0.1.tar.gz
Windows binaries: r-prerel: PAGFL_1.0.1.zip, r-release: PAGFL_1.0.1.zip, r-oldrel: PAGFL_1.0.1.zip
macOS binaries: r-prerel (arm64): PAGFL_1.0.1.tgz, r-release (arm64): PAGFL_1.0.1.tgz, r-oldrel (arm64): PAGFL_1.0.1.tgz, r-prerel (x86_64): PAGFL_1.0.1.tgz, r-release (x86_64): PAGFL_1.0.1.tgz
Old sources: PAGFL archive

Linking:

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