rfUtilities (version 2.1-3)

rf.significance: Random Forest model significance test

Description

Performs significance test for classification and regression Random Forests models.

Usage

rf.significance(x, xdata, q = 0.99, p = 0.05, nperm = 999, ...)

Arguments

x

randomForest class object

xdata

Independent variables (x) used in model

q

Quantile threshold to test classification models

p

p-value to test for significance in regression models

nperm

Number of permutations

...

Additional Random Forests arguments

Value

A list class object with the following components:

For Regression problems:

• "RandRsquare" Vector of random R-square values

• "Rsquare" The R-square of the "true" model

• "Accept" model significant at specified p-value (TRUE/FALSE)

• "TestQuantile" Quantile threshold used in significance plot

• "pValueThreshold" Specified p-value

• "pValue" p-values of randomizations

• "nPerm" Number of permutations

For Classification problems:

• "RandOOB" Vector of random out-of-bag (OOB) values

• "RandMaxError" Maximum error of randomizations

• "test.OOB" Error if the "true" model

• "Accept" Is the model significant at specified p-value (TRUE/FALSE)

• "TestQuantile" Quantile threshold used in significance plot

• "pValueThreshold" Specified p-value

• "pValue" p-values of randomizations

• "nPerm" Number of permutations

References

Murphy M.A., J.S. Evans, and A.S. Storfer (2010) Quantify Bufo boreas connectivity in Yellowstone National Park with landscape genetics. Ecology 91:252-261

Evans J.S., M.A. Murphy, Z.A. Holden, S.A. Cushman (2011). Modeling species distribution and change using Random Forests CH.8 in Predictive Modeling in Landscape Ecology eds Drew, CA, Huettmann F, Wiersma Y. Springer

Examples

# NOT RUN {
# Regression
require(randomForest)
  set.seed(1234)	
    data(airquality)
    airquality <- na.omit(airquality)
 ( rf.mdl <- randomForest(x=airquality[,2:6], y=airquality[,1]) )
   ( rf.perm <- rf.significance(rf.mdl, airquality[,2:6], nperm=99, ntree=501) )
 
# Classification
require(randomForest)
  set.seed(1234)	
    data(iris)
      iris$Species <- as.factor(iris$Species) 
 ( rf.mdl <- randomForest(iris[,1:4], iris[,"Species"], ntree=501) )
   ( rf.perm <- rf.significance(rf.mdl, iris[,1:4], nperm=99, ntree=501) ) 
# }
# NOT RUN {
# }