LambertW: Analyze and Gaussianize skewed, heavy-tailed data

The Lambert W framework is a new generalized way to analyze skewed, heavy-tailed data. Lambert W random variables (RV) are based on an input/output framework where the input is a RV X with distribution F(x), and the output Y = func(X) has similar properties as X (but slightly skewed or heavy-tailed). Then this transformed RV Y has a Lambert W x F distribution - for details see References. This package contains functions to perform a Lambert W analysis of skewed and heavy-tailed data: data can be simulated, parameters can be estimated from real world data, quantiles can be computed, and results plotted/printed in a 'nice' way. Probably the most important function is 'Gaussianize', which works the same way as the R function 'scale' but actually makes your data Gaussian. An optional modular toolkit implementation allows users to define their own Lambert W x 'my favorite distribution' and use it for their analysis.

Version: 0.2.9.9
Depends: moments, gsl, MASS, nortest, maxLik
Published: 2012-01-08
Author: Georg M. Goerg
Maintainer: Georg M. Goerg <gmg at stat.cmu.edu>
License: GPL (≥ 2)
URL: http://www.stat.cmu.edu/~gmg http://arxiv.org/abs/0912.4554 http://arxiv.org/abs/1010.2265
In views: Distributions
CRAN checks: LambertW results

Downloads:

Package source: LambertW_0.2.9.9.tar.gz
MacOS X binary: LambertW_0.2.9.9.tgz
Windows binary: LambertW_0.2.9.9.zip
Reference manual: LambertW.pdf
Old sources: LambertW archive