Probabilistic Models to Analyze and Gaussianize Heavy-Tailed, Skewed Data.
LambertW R package
This is the github repo for the LambertW R package hosted on CRAN. For any changes after the official version, see the commit history and here.
Installation & usage
To install LambertW run
install.packages("LambertW")
citation("LambertW")
See ?LambertW
for examples on how to use the LambertW package.
There is also an R vignette on CRAN with a brief tutorial on the main functionalities.
Python implementation
See https://github.com/gmgeorg/pylambertw for the Python equivalent of the LambertW package.
Tutorials & posts
See cross-validated / stackoverflow for a variety of LambertW posts on how to normalize/Gaussianize data and model skewed/heavy-tailed distributions.
References
Georg M. Goerg (2011): Lambert W random variables - a new family of generalized skewed distributions with applications to risk estimation. Annals of Applied Statistics 3(5). 2197-2230.
Georg M. Goerg (2014): The Lambert Way to Gaussianize heavy-tailed data with the inverse of Tukey's h transformation as a special case. The Scientific World Journal.