rfUtilities (version 2.0-0)

rf.partial.prob: Random Forest probability scaled partial dependency plots

Description

Produces partial dependency plots with probability distribution based on scaled margin distances.

Usage

rf.partial.prob(x, pred.data, xname, which.class, w, prob = TRUE, plot = TRUE, smooth = FALSE, raw = FALSE, rug = FALSE, n.pt, xlab, ylab, main, ...)

Arguments

x

Object of class randomForest

pred.data

Training data.frame used for constructing the plot,

xname

Name of the variable for calculating partial dependence

which.class

The class to focus on

w

Weights to be used in averaging (if not supplied, mean is not weighted)

prob

Scale distances to probabilities

plot

Plot results (TRUE/FALSE)

smooth

Apply spline.smooth to y

raw

Plot unsmoothed values

rug

Draw hash marks on plot representing deciles of x

n.pt

Number of points on the grid for evaluating partial dependence.

xlab

x-axis plot label

ylab

y-axis plot label

main

Plot label for main

...

Additional graphical parameters passed to plot

Value

A list class object with fit x,y. If smooth=TRUE y represents smoothed scaled margin distance values

References

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

Baruch-Mordo, S., J.S. Evans, J. Severson, J. D. Naugle, J. Kiesecker, J. Maestas, & M.J. Falkowski (2013) Saving sage-grouse from the trees: A proactive solution to reducing a key threat to a candidate species Biological Conservation 167:233-241

Examples

require(randomForest)
  data(iris)
    iris.rf <- randomForest(iris[,1:4], iris[,5])
      par(mfrow=c(2,2))
        for(i in names(iris)[1:4]) {     
          rf.partial.prob(iris.rf, iris, i, "setosa", smooth=TRUE, raw=TRUE, rug=FALSE)
         }