qerror {forestError}

Estimated conditional prediction error quantile functions

Description

Returns quantiles of the estimated conditional error distribution associated with each test prediction.

Usage

qerror(p, xs)

Arguments

p

A vector of probabilities.

xs

A vector of the indices of the test observations for which the conditional error quantiles are desired. Defaults to all test observations given in the call of quantForestError.

Details

This function is only defined as output of the quantForestError function. It is not exported as a standalone function. See the example.

Value

If either p or xs has length one, then a vector is returned with the desired quantiles. If both have length greater than one, then a data.frame of quantiles is returned, with rows corresponding to the inputted xs and columns corresponding to the inputted p.

Author(s)

Benjamin Lu <[e-mail]>; Johanna Hardin <[e-mail]>

See Also

quantForestError

Examples

# load data
data(airquality)

# remove observations with missing predictor variable values
airquality <- airquality[complete.cases(airquality), ]

# get number of observations and the response column index
n <- nrow(airquality)
response.col <- 1

# split data into training and test sets
train.ind <- sample(1:n, n * 0.9, replace = FALSE)
Xtrain <- airquality[train.ind, -response.col]
Ytrain <- airquality[train.ind, response.col]
Xtest <- airquality[-train.ind, -response.col]
Ytest <- airquality[-train.ind, response.col]

# fit random forest to the training data
rf <- randomForest::randomForest(Xtrain, Ytrain, nodesize = 5,
                                 ntree = 500,
                                 keep.inbag = TRUE)

# estimate conditional error distribution functions
output <- quantForestError(rf, Xtrain, Xtest,
                           what = c("p.error", "q.error"))

# get the 0.25 and 0.8 quantiles of the error distribution
# associated with each test prediction
output$qerror(c(0.25, 0.8))

# same as above but only for the first three test observations
output$qerror(c(0.25, 0.8), 1:3)