screen.wgtd.elasticnet: Weighted regularized regression screening algorithm

In saraemoore/SLScreenExtra: A Collection of Additional Feature Selection Algorithms and Utilities for SuperLearner

Weighted regularized regression screening algorithm

Description

Performs feature selection via cv.glmnet.

Usage

screen.wgtd.elasticnet(
  Y,
  X,
  family,
  obsWeights,
  id,
  alpha = 0.5,
  k = 2,
  nfolds = 10,
  nlambda = 100,
  ...
)

screen.wgtd.lasso(..., alpha = 1)

Arguments

Y	Outcome (numeric vector). See SuperLearner for specifics.
X	Predictor variable(s) (data.frame or matrix). See SuperLearner for specifics.
family	Error distribution to be used in the model: gaussian or binomial. See SuperLearner for specifics.
obsWeights	Optional numeric vector of observation weights. See SuperLearner for specifics.
id	Cluster identification variable. Currently unused.
alpha	The elasticnet mixing parameter. Default for screen.wgtd.elasticnet is (arbitrarily) 0.5. Forced to 1 for screen.wgtd.lasso. See glmnet for specifics.
k	Minimum number of features to select. Only used if fewer than this number of features are selected using the optimal lambda value chosen via cross-validation.
nfolds	Number of cross-validation folds to use when choosing optimal lambda. Default is 10. See cv.glmnet for specifics.
nlambda	Number of lambda values to try. Default is 100. See glmnet for specifics.
...	Currently unused.

Value

A logical vector with length equal to ncol(X)

Examples

# based on example in SuperLearner package
set.seed(1)
n <- 100
p <- 20
X <- matrix(rnorm(n*p), nrow = n, ncol = p)
X <- data.frame(X)
Y <- X[, 1] + sqrt(abs(X[, 2] * X[, 3])) + X[, 2] - X[, 3] + rnorm(n)
obsWeights <- 1/runif(n)
screen.wgtd.elasticnet(Y, X, gaussian(), obsWeights, seq(n), k = 3)
screen.wgtd.lasso(Y, X, gaussian(), obsWeights, seq(n), k = 3)

saraemoore/SLScreenExtra documentation built on Nov. 4, 2023, 9:31 p.m.