README.md

In xu1912/EPSLASSO: Sparse Regression for data under extreme phenotype sampling

EPSLASSO

An R package for regression analysis of data from extreme sampling. Both of the low-dimensional (n>p) and high-dimensional (n<p) methods are available.

Contact

Chao Xu [e-mail]

Installation

If the R package devtools can be installed, then try:

library("devtools")

install_github("xu1912/EPSLASSO")

If devtools is not available, then have to download the source code and install local using zip (for Windows) or tar.gz (for Linux) file. Before the installation, check the required packages:

  glmnet(>= 2.0-5),

  doParallel (>= 1.0.1),

  flare (>= 1.5.0),

  foreach (>= 1.4.0),

  clime (>= 0.4.1),

  mvtnorm (>= 1.0-3),

  methods (>= 3.2).

Tutorial

The main function of this package is epsLasso(X, Y, c1, c2), where X is the matrix of predictors with each row represents an individual; Y is the phenotypic value; c1 is the right censored phenotypic value; c2 is the left censored phenotypic value. (to be continued)

xu1912/EPSLASSO documentation built on May 10, 2022, 11:23 a.m.