enet: enet

View source: R/regression_enet.R

enetR Documentation

enet

Description

enet computes the elastic net estimator using the cyclic co-ordinate descent (CCD) algorithm.

Usage

enet(y, X, beta, lambda, alpha = 1, printitn = 0, itermax = 1000)

Arguments

y

: (numeric) data vector of size N x 1 (output, respones) if the intercept is in the model, then y needs to be centered.

X

: (numeric) data matrix of size N x p (input, features) Columns are assumed to be standardized (i.e., norm(X(:,j))=1) as well as centered (if intercept is in the model).

beta

: (numeric) regression vector for initial start for CCD algorithm

lambda

: (numeric) a postive penalty parameter value

alpha

: (numeric) elastic net tuning parameter in the range [0,1]. If not given then use alpha = 1 (Lasso)

printitn:

print iteration number (default = 0, no printing)

Value

beta : (numeric) the regression coefficient vector

iter : (numeric) # of iterations


Mufabo/Rrobustsp documentation built on June 11, 2022, 10:41 p.m.