varSelRanger: Variable selection using recursive feature elimination.

In SimonLarsen/varSelRanger: Variable selection using the ranger random forest package

Description

Models are evaluated based on the OOB error rate.

Usage

varSelRanger(data, dependent.variable.name, formula = NULL, frac = 0.2,
  min.vars = 1, importance = "impurity", ...)

Arguments

data	Training data of class data.frame, matrix or dgCMatrix.
dependent.variable.name	Name of the dependent variable. For survival forests this is the time variable.
formula	Object of class formula or character describing the model to fit. Interaction terms supported only for numerical variables. (Formula interface not yet supported in varSelRanger).
frac	Fraction of top variables to drop after each iteration.
min.vars	Minimum number of variables allowed in a model.
importance	Variable importance mode. One of 'none', 'impurity', 'impurity_corrected', 'permutation'. The 'impurity' measure is the Gini index for classification, the variance of the responses for regression and the sum of test statistics (see splitrule) for survival.
...	Further arguments to be passed to

SimonLarsen/varSelRanger documentation built on April 2, 2019, 3:07 p.m.