rForest: Random Forest classfication in statTarget

Description Usage Arguments Value Author(s) References Examples

View source: R/statTarget_rforest.R

Description

rForest provides the Breiman's random forest algorithm for classification and permutation-based variable importance measures (PIMP-algorithm).

Usage

1
2
rForest(file,ntree = 100,times = 100, gDist = TRUE,
seed = 123,...)

Arguments

file

An data frame or 'Stat File' from statTarget software.

ntree

Number of trees to grow. This should not be set to too small a number, to ensure that every input row gets predicted at least a few times.

times

The number of permutations for permutation-based variable importance measures.

gDist

If gDist is TRUE the null importance distributions are approximated with Gaussian distributions else with empirical cumulative distributions.

seed

For the same set of random variables and reproducible results.

...

A generic function in randomForest package

Value

Objects Two objects from statTarget_rForest (1. randomForest,rfModel; 2. PIMPresult, pimpModel)

VarImp The original Gini importance

PerVarImp A matrix, where the permuted VarImp measures for the predictor variable.

p-value The probability of observing the original VarImp or a larger value, given the fitted null importance distribution.

p.ks.test The p-values of the Kolmogorov-Smirnov Tests for each row PerVarImp.

Author(s)

Hemi Luan, hemi.luan@gmail.com

References

Altmann A.,Tolosi L.,Sander O. and Lengauer T. (2010) Permutation importance: a corrected feature importance measure, Bioinformatics 26 (10), 1340-1347.

Ender Celik. (2015) vita: Variable Importance Testing Approaches. R package version 1.0.0 https://CRAN.R-project.org/package=vita

Examples

1
2
3
4
datpath <- system.file('extdata',package = 'statTarget')
statFile <- paste(datpath,'data_example.csv', sep='/')
getFile <- read.csv(statFile,header=TRUE)
rFtest <- rForest(getFile,ntree = 10,times = 5)

13479776/statTarget documentation built on Aug. 14, 2020, 1:58 p.m.