ggRandomForests (version 1.0.0)

gg_interaction.ggRandomForests: Minimal Depth Variable Interaction data object (randomForestSRC::find.interaction).

Description

Basically, this function adds attributes to the results of running randomForestSRC::find.interaction on an randomForestSRC::rfsrc random forest. If passed a randomForestSRC::rfsrc object, gg_interaction first runs the randomForestSRC::find.interaction function with all optional arguments.

Usage

gg_interaction.ggRandomForests(object, ...)

Arguments

object

a randomForestSRC::rfsrc object or the output from the randomForestSRC::find.interaction function call.

...

optional extra arguments passed to find.interaction.

References

Ishwaran H. (2007). Variable importance in binary regression trees and forests, Electronic J. Statist., 1:519-537.

Ishwaran H., Kogalur U.B., Gorodeski E.Z, Minn A.J. and Lauer M.S. (2010). High-dimensional variable selection for survival data. J. Amer. Statist. Assoc., 105:205-217.

Ishwaran H., Kogalur U.B., Chen X. and Minn A.J. (2011). Random survival forests for high-dimensional data. Statist. Anal. Data Mining, 4:115-132.

See Also

plot.gg_interaction randomForestSRC::rfsrc randomForestSRC::find.interaction randomForestSRC::max.subtree randomForestSRC::var.select randomForestSRC::vimp

Examples

## Examples from randomForestSRC package...
## ------------------------------------------------------------
## find interactions, classification setting
## ------------------------------------------------------------
## iris.obj <- rfsrc(Species ~., data = iris)
## TODO: VIMP interactions not handled yet....
## find.interaction(iris.obj, method = "vimp", nrep = 3)
## iris_interaction <- find.interaction(iris.obj)
data(iris_interaction, package="ggRandomForests")
gg_int <- gg_interaction(iris_interaction)

plot(gg_int, x_var="Petal.Width")
plot(gg_int, x_var="Petal.Length")

## ------------------------------------------------------------
## find interactions, regression setting
## ------------------------------------------------------------
## airq.obj <- rfsrc(Ozone ~ ., data = airquality)
##
## TODO: VIMP interactions not handled yet....
## find.interaction(airq.obj, method = "vimp", nrep = 3)
## airq_interaction <- find.interaction(airq.obj)
data(airq_interaction, package="ggRandomForests")
gg_int <- gg_interaction(airq_interaction)

plot(gg_int, x_var="Temp")
plot(gg_int, x_var="Solar.R")

## ------------------------------------------------------------
## find interactions, survival setting
## ------------------------------------------------------------
## data(pbc, package = "randomForestSRC")
## pbc.obj <- rfsrc(Surv(days,status) ~ ., pbc, nsplit = 10)
## pbc_interaction <- find.interaction(pbc.obj, nvar = 8)
data(pbc_interaction, package="ggRandomForests")
gg_int <- gg_interaction(pbc_interaction)

plot(gg_int, x_var="bili")
plot(gg_int, x_var="copper")