ReproStat: Reproducibility Diagnostics for Statistical Modeling

Tools for diagnosing the reproducibility of statistical model outputs under data perturbations. Implements bootstrap, subsampling, and noise-based perturbation schemes and computes coefficient stability, p-value stability, selection stability, prediction stability, and a composite reproducibility index on a 0 to 100 scale. Includes cross-validation ranking stability for model comparison and visualization utilities. Optional 'backends' support robust M-estimation ('MASS') and penalized regression ('glmnet'). Bootstrap perturbation follows 'Efron' and 'Tibshirani' (1993, ISBN:9780412042317); selection stability follows 'Meinshausen' and 'Buhlmann' (2010) <doi:10.1111/j.1467-9868.2010.00740.x>; reproducibility framework follows 'Peng' (2011) <doi:10.1126/science.1213847>.

Version: 0.1.2
Imports: stats, graphics
Suggests: testthat (≥ 3.0.0), knitr, rmarkdown, MASS (≥ 7.3), glmnet (≥ 4.0), ggplot2 (≥ 3.0)
Published: 2026-03-30
DOI: 10.32614/CRAN.package.ReproStat (may not be active yet)
Author: Gideon Nti Boateng [aut, cre]
Maintainer: Gideon Nti Boateng <gidiboateng200 at gmail.com>
BugReports: https://github.com/ntiGideon/ReproStat/issues
License: GPL (≥ 3)
URL: https://ntigideon.github.io/ReproStat/, https://github.com/ntiGideon/ReproStat
NeedsCompilation: no
Materials: README, NEWS
CRAN checks: ReproStat results

Documentation:

Reference manual: ReproStat.html , ReproStat.pdf
Vignettes: Introduction to ReproStat (source, R code)
Backend Guide (source, R code)
Interpreting ReproStat Outputs (source, R code)
Workflow Patterns (source, R code)

Downloads:

Package source: ReproStat_0.1.2.tar.gz
Windows binaries: r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries: r-release (arm64): ReproStat_0.1.2.tgz, r-oldrel (arm64): not available, r-release (x86_64): ReproStat_0.1.2.tgz, r-oldrel (x86_64): ReproStat_0.1.2.tgz

Linking:

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