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. (Source: http://cran.r-project.org/web/packages)
Keywords for this software
References in zbMATH (referenced in 2 articles , 1 standard article )
Showing results 1 to 2 of 2.
- Witkovský, Viktor; Wimmer, Gejza; Duby, Tomy: Logarithmic Lambert $W \times \mathcalF$ random variables for the family of chi-squared distributions and their applications (2015)
- Goerg, Georg M.: Lambert $W$ random variables -- a new family of generalized skewed distributions with applications to risk estimation (2011)