enetCV: M-fold cross-validation function for Elastic net

In singha53/ssenet: Semi-supervised implementation of glmnet

Description

Estimate test error of elastic net panel

Usage

enetCV(
  X,
  y,
  alpha,
  lambda,
  folds,
  progressBar,
  family,
  filter,
  topranked,
  keepVar,
  weights
)

Arguments

X	nxp matrix - training dataset
y	factor - response variable
alpha	= 1 (lasso), alpha = 0 (ridge), 0 < alpha < 1 (elastic net penalty)
lambda	= strength of elastic net penalty
folds	caret list specifying the indicies in the different folds
progressBar	a boolean that specifies whether a progress bar should be displayed or not
family	"binomial" or "multinomial"
filter	= "none" or "p.value"
topranked	= 50 (top number of features to select and build a classifier)
keepVar	- names of specific variable to keep in model
weights	- observational weights; default to 1

singha53/ssenet documentation built on March 17, 2020, 4:41 a.m.