Variable Selection for Highly Correlated Predictors
It proposes a novel variable selection approach taking into account the correlations that may exist between the predictors of the design matrix in a high-dimensional linear model. Our approach consists in rewriting the initial high-dimensional linear model to remove the correlation between the predictors and in applying the generalized Lasso criterion. For further details we refer the reader to the paper <arXiv:2007.10768> (Zhu et al., 2020).
The DESCRIPTION file:
| Package: | WLasso |
| Type: | Package |
| Title: | Variable Selection for Highly Correlated Predictors |
| Version: | 1.0 |
| Date: | 2020-08-07 |
| Authors@R: | c(person("Wencan", "Zhu", email = "wencan.zhu@agroparistech.fr", role = c("aut", "cre")), person("Celine","Levy-Leduc", email="celine.levy-leduc@agroparistech.fr", role = "ctb"), person("Nils", "Ternes", email="Nils.Ternes@sanofi.com", role = "ctb")) |
| Author: | Wencan Zhu [aut, cre], Celine Levy-Leduc [ctb], Nils Ternes [ctb] |
| Maintainer: | Wencan Zhu <wencan.zhu@agroparistech.fr> |
| Description: | It proposes a novel variable selection approach taking into account the correlations that may exist between the predictors of the design matrix in a high-dimensional linear model. Our approach consists in rewriting the initial high-dimensional linear model to remove the correlation between the predictors and in applying the generalized Lasso criterion. For further details we refer the reader to the paper <arXiv:2007.10768> (Zhu et al., 2020). |
| License: | GPL-2 |
| Imports: | Matrix, genlasso, tibble, MASS, ggplot2 |
| VignetteBuilder: | knitr |
| Suggests: | knitr, markdown |
| NeedsCompilation: | no |
| Packaged: | 2020-08-07 12:09:08 UTC; wencan |
| Depends: | R (>= 3.5.0) |
Index of help topics:
Sigma_Estimation Estimation of the correlation matrix
WLasso-package Variable Selection for Highly Correlated
Predictors
Whitening_Lasso Whitening Lasso
X Example of a design matrix of a linear model
Y Example of a response variable of a linear
model.
top Thresholding to zero of the smallest values
top_thresh Thresholding to a given threshold of the
smallest valuesThis package consists of four functions: "Sigma_Estimation.R", "top.R", "top_thresh.R" and "Whitening_Lasso.R". For further information on how to use these functions, we refer the reader to the vignette of the package.
Wencan Zhu [aut, cre], Celine Levy-Leduc [ctb], Nils Ternes [ctb]
Maintainer: Wencan Zhu <wencan.zhu@agroparistech.fr>
W. Zhu, C. Levy-Leduc, N. Ternes. "A variable selection approach for highly correlated predictors in high-dimensional genomic data". arXiv:2007.10768.
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.