osktnorm: A Moment-Targeting Normality Transformation Based on Tukey g-h Distribution

Implements a moment-targeting normality transformation based on the simultaneous optimization of Tukey g-h distribution parameters. The method is designed to minimize both asymmetry (skewness) and excess peakedness (kurtosis) in non-normal data by mapping it to a standard normal distribution Cebeci et al (2026) <doi:10.3390/sym18030458>. Optimization is performed by minimizing an objective function derived from the Anderson-Darling goodness-of-fit statistic with Stephens's correction factor, utilizing the L-BFGS-B algorithm for robust parameter estimation. This approach provides an effective alternative to power transformations like Box-Cox and Yeo-Johnson, particularly for data requiring precise tail-behavior adjustment.

Version:	1.1.2
Depends:	R (≥ 4.1.0)
Imports:	Rcpp, stats, doParallel, foreach, parallel, groupcompare
LinkingTo:	Rcpp
Suggests:	knitr, rmarkdown, readxl, writexl
Published:	2026-03-17
DOI:	10.32614/CRAN.package.osktnorm
Author:	Zeynel Cebeci [aut, cre], Figen Ceritoglu [aut], Melis Celik Guney [aut], Adnan Unalan [aut]
Maintainer:	Zeynel Cebeci <cebeciz at gmail.com>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation:	yes
Citation:	osktnorm citation info
Materials:	README
CRAN checks:	osktnorm results

Documentation:

Reference manual:	osktnorm.html , osktnorm.pdf
Vignettes:	OSKT: Normality Transformation via Optimized Skewness and Kurtosis (source, R code) A Practical Example for Working with Datasets (source, R code)

Downloads:

Package source:	osktnorm_1.1.2.tar.gz
Windows binaries:	r-devel: osktnorm_1.1.2.zip, r-release: osktnorm_1.1.2.zip, r-oldrel: not available
macOS binaries:	r-release (arm64): osktnorm_1.1.2.tgz, r-oldrel (arm64): not available, r-release (x86_64): osktnorm_1.1.2.tgz, r-oldrel (x86_64): not available

Linking:

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