WLogit-package {WLogit}

Variable Selection in High-Dimensional Logistic Regression Models using a Whitening Approach

Description

It proposes a novel variable selection approach in classification problem that takes into account the correlations that may exist between the predictors of the design matrix in a high-dimensional logistic model. Our approach consists in rewriting the initial high-dimensional logistic model to remove the correlation between the predictors and in applying the generalized Lasso criterion.

Details

The DESCRIPTION file:

Package:	WLogit
Type:	Package
Title:	Variable Selection in High-Dimensional Logistic Regression Models using a Whitening Approach
Version:	2.1
Date:	2023-07-17
Author:	Wencan Zhu
Maintainer:	Wencan Zhu <[e-mail]>
Description:	It proposes a novel variable selection approach in classification problem that takes into account the correlations that may exist between the predictors of the design matrix in a high-dimensional logistic model. Our approach consists in rewriting the initial high-dimensional logistic model to remove the correlation between the predictors and in applying the generalized Lasso criterion.
License:	GPL-2
Imports:	cvCovEst, genlasso, tibble, MASS, ggplot2, Matrix, glmnet, corpcor
VignetteBuilder:	knitr
Suggests:	knitr
Depends:	R (>= 3.5.0)
NeedsCompilation:	no
Packaged:	2023-07-17 07:06:43 UTC; mmip

Index of help topics:

CalculPx                Calculate the class-conditional probabilities.
CalculWeight            Calculate the weight
Refit_glm               Refit the logistic regression with chosen
                        variables
Thresholding            Thresholding on a vector
WLogit-package          Variable Selection in High-Dimensional Logistic
                        Regression Models using a Whitening Approach
WhiteningLogit          Variable selection in high-dimensional logistic
                        regression models using a whitening approach
WorkingResp             Calculate the working response
X                       Example of a design matrix of a logistic model
beta                    True coefficients in the esample.
test                    WLogit output
top                     Thresholding to zero of the smallest values
top_thresh              Thresholding to a given threshold of the
                        smallest values
y                       Example of a binary response variable of a
                        logistic model.

Further information is available in the following vignettes:

Vignettes	WLogit package (source, pdf)

This package consists of functions: "WhiteningLogit", "CalculPx", "CalculWeight", "Refit_glm", "top", "top_thresh", "WorkingResp", and "Thresholding". For further information on how to use these functions, we refer the reader to the vignette of the package.

Author(s)

Wencan Zhu

Maintainer: Wencan Zhu <[e-mail]>

References

W. Zhu, C. Levy-Leduc, N. Ternes. "Variable selection in high-dimensional logistic regression models using a whitening approach". (2022)