permtestImp: Testing Importance of All Variables

In tim-coleman/RFtest: Scalable and Efficient Hypothesis Tests for Random Forests

Description

This method runs the variable importance procedure (MSE_Test) on each variable in the data frame, marginally, with an available plot method.

Usage

permtestImp(X, y, X.test = FALSE, y.test = FALSE,
  single_forest = T, NTest = nrow(X.test), Nbtree = 30,
  verbose = F, keep_forest = F, base.learner = "rpart",
  mtry = ncol(X), p = 0.5, ...)

Arguments

X	Data frame of covariates - the training data.
y	Response vector. Currently only numeric responses (regression) are supported.
X.test	Covariates of the test set with which the MSE is calculated.
y.test	Responses in the test set with which the MSE is calculated.
base.learner	One of "rpart", "ctree", "rtree", or "lm". Base model to be used in the bagging.
single_forest	Logical. If TRUE, then all variables are compared against a single original forest.
NTest	If X.test, y.test are not specified, this number of test points are drawn at random from X, y.
Nbtree	How many trees should be used for each variable. Can either be a single number, or a vector of length ncol(X), with each entry corresponding to the number of trees used for testing each column.
verbose	Logical. Should a progress tracker be output in the console?
keep_forest	Logical. Should the original random forest be returned?
mtry	The mtry parameter used in random forest models.
p	Subsample size exponent, see MSE_Test.
...	Additional arguments to be passed to MSE_Test.

Value

Returns an object of S4 class permtestImp with the following values

ImportanceTable	A data.frame tabulating each variable and its importance.
call	The original call, as provided by match.call()

Examples

set.seed(367116)
N <- 1250
Nvar <- 5
name_vec <- paste("X", 1:(2*Nvar), sep = "")

# training data:
X <- data.frame(replicate(Nvar, runif(N)),
                replicate(Nvar, cut(runif(N), 3,
                                      labels = as.character(1:3))))
names(X) <- name_vec

## Example where response is unrelated to any features
pm.null <- permtestImp(X = X, y = rnorm(N),
                         single_forest = F, base.learner = "rtree",
                        mtry = 3, NTest = 100, Nbtree = 100, B = 1000,
                       p = .75, verbose = T)

## Plotting the Importance Output
plot(pm.null)

tim-coleman/RFtest documentation built on March 10, 2020, 12:28 p.m.