holdout.rf: Helper function for variable selection using Vita approach.

In silkeszy/Pomona: Identification of relevant variables in omics data sets using Random Forests

Helper function for variable selection using Vita approach.

Description

This function calculates a modified version of the permutation importance using two cross-validation folds (holdout folds) as described in Janitza et al. (2015). Note that this function is a reimplementation of the holdoutRF function in the R package ranger.

Usage

holdout.rf(
  x,
  y,
  ntree = 500,
  mtry.prop = 0.2,
  nodesize.prop = 0.1,
  no.threads = 1,
  type = "regression",
  importance = importance
)

Arguments

x	matrix or data.frame of predictor variables with variables in columns and samples in rows (Note: missing values are not allowed).
y	vector with values of phenotype variable (Note: will be converted to factor if classification mode is used).
ntree	number of trees.
mtry.prop	proportion of variables that should be used at each split.
nodesize.prop	proportion of minimal number of samples in terminal nodes.
no.threads	number of threads used for parallel execution.
type	mode of prediction ("regression", "classification" or "probability").
importance	See ranger for details.

Value

Hold-out random forests with variable importance

References

Janitza, S., Celik, E. & Boulesteix, A.-L., (2015). A computationally fast variable importance test for random forest for high dimensional data, Technical Report 185, University of Munich, https://epub.ub.uni-muenchen.de/25587.

silkeszy/Pomona documentation built on March 31, 2022, 11:13 p.m.