Learn R Programming

LassoNet

The LassoNet package is the implementation of 3CoSE algorithm proposed in the article [1] and [2].

References

[1] Weber, Matthias and Schumacher, Martin and Binder, Harald, Regularized Regression Incorporating Network Information: Simultaneous Estimation of Covariate Coefficients and Connection Signs (June 28, 2014). Tinbergen Institute Discussion Paper 14-089/I. Available at SSRN: https://ssrn.com/abstract=2466289 or http://dx.doi.org/10.2139/ssrn.2466289

[2] Weber, Matthias and Striaukas, Jonas and Schumacher, Martin and Binder, Harald, Network-Constrained Covariate Coefficient and Connection Sign Estimation (June 24, 2018). CORE Discussion Paper 2018/18 -OR- Bank of Lithuania Discussion Paper No 8/2018. Available at SSRN: https://ssrn.com/abstract=3211163 or http://dx.doi.org/10.2139/ssrn.3211163

Copy Link

Version

Install

install.packages('LassoNet')

Monthly Downloads

156

Version

0.8.3

License

GPL (>= 2)

Maintainer

Jonas Striaukas

Last Published

January 19th, 2020

Functions in LassoNet (0.8.3)

matrix.M.update

Updates connection sign matrix.
soft.thresh

Soft thresholding operator.
lasso.net.fixed

Estimates coefficients over the grid values of penalty parameters.
lasso.net.grid

Estimates coefficients and connection signs over the grid of values of penalty parameters \(\lambda\)1 and \(\lambda\)2.
fastols

Fast least squares estimate.
get.xi

Updates the estimates of the connection signs by running mini OLS models.
mat.to.laplacian

Computes Laplacian matrix.
get.BxBy

Computes decomposition elements.
LassoNet-package

LassoNet: package for 3CoSE algorithm.
betanew_lasso_cpp

C++ subroutine that updates \(\beta\) coefficients.
beta.update.net

Updates \(\beta\) coefficients.
get.signs.M

Vetorizes connection sign matrix.