ADMMnet: Regularized Model with Selecting the Number of Non-Zeros

Fit linear and cox models regularized with net (L1 and Laplacian), elastic-net (L1 and L2) or lasso (L1) penalty, and their adaptive forms, such as adaptive lasso and net adjusting for signs of linked coefficients. In addition, it treats the number of non-zero coefficients as another tuning parameter and simultaneously selects with the regularization parameter. The package uses one-step coordinate descent algorithm and runs extremely fast by taking into account the sparsity structure of coefficients.

AuthorXiang Li, Shanghong Xie, Donglin Zeng and Yuanjia Wang
Date of publication2015-12-12 09:41:54
MaintainerXiang Li <xl2473@cumc.columbia.edu>
LicenseGPL (>= 2)
Version0.1

View on CRAN

Files in this package

ADMMnet
ADMMnet/inst
ADMMnet/inst/include
ADMMnet/inst/include/ADMMnet.h
ADMMnet/inst/include/ADMMnet_RcppExports.h
ADMMnet/src
ADMMnet/src/ADMMnet.cpp
ADMMnet/src/RcppExports.cpp
ADMMnet/NAMESPACE
ADMMnet/R
ADMMnet/R/net_cox.R ADMMnet/R/print.ADMMnet.R ADMMnet/R/RcppExports.R ADMMnet/R/ADMMini.R ADMMnet/R/net_lm.R ADMMnet/R/ADMMnet.R
ADMMnet/MD5
ADMMnet/DESCRIPTION
ADMMnet/man
ADMMnet/man/ADMMnet-package.Rd ADMMnet/man/print.ADMMnet.Rd ADMMnet/man/ADMMnet.Rd

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.