screen.glmnet.fix: Lasso screener selects pre-specified variables

In sl-bergquist/SLscreeners: Screening Wrappers for Super Learner

Description

Lasso screener for SuperLearner() that always retains specified variables and passes approximately nVar variables to SuperLearner(). When the number of non-zero coefficients exceeds nVar, a larger value of the regularization parameter lambda is chosen to select a smaller set of variables that excludes the ties.

Usage

screen.glmnet.fix(Y, X, family, alpha = 1, minscreen = 2, nVar = 10,
  nfolds = 10, nlambda = 100, fixed.var.index = var.index, ...)

Arguments

Y	outcome variable (specified in SuperLearner())
X	data frame
nVar	number of non-zero variables to be selected
var.index	indices of variables to always be included by the screener

See Also

See glmnet for additional details on implementing lasso

Examples

If you do not know the indices of the variables you always want to include, 
  you can get them from the variable name, where newdat is the dataframe: 
  var.index <- c(which(colnames(newdat)=="tcls14"), 
                 which(colnames(newdat)=="tcls251"))

sl-bergquist/SLscreeners documentation built on Dec. 2, 2019, 1:29 a.m.