Estimates the prediction error distribution parameters requested in the input
to quantForestError.
estimateErrorParams(
train_nodes,
test_nodes,
mspewhat,
biaswhat,
intervalwhat,
pwhat,
qwhat,
mcrwhat,
alpha,
n.test
)A data.table indicating which out-of-bag prediction
errors are in each terminal node of each tree in the random forest. It
must be formatted like the output of the findOOBErrors function.
A data.table indicating which test observations are
in each terminal node of each tree in the random forest. It must be
formatted like the output of the findTestPreds function.
A boolean indicating whether to estimate conditional MSPE.
A boolean indicating whether to estimate conditional bias.
A boolean indicating whether to estimate conditional prediction intervals.
A boolean indicating whether to estimate the conditional prediction error CDFs.
A boolean indicating whether to estimate the conditional prediction error quantile functions.
A boolean indicating whether to estimate the conditional misclassification rate.
A vector of type-I error rates desired for the conditional prediction
intervals; required if intervalwhat is TRUE.
The number of test observations.
A data.frame with one or more of the following columns:
The random forest predictions of the test observations
The estimated conditional mean squared prediction errors of the random forest predictions
The estimated conditional biases of the random forest predictions
The estimated lower bounds of the conditional alpha-level prediction intervals for the test observations
The estimated upper bounds of the conditional alpha-level prediction intervals for the test observations
The estimated conditional misclassification rate of the random forest predictions
In addition, one or both of the following functions:
The estimated cumulative distribution functions of the conditional error distributions associated with the test predictions
The estimated quantile functions of the conditional error distributions associated with the test predictions
This function is for internal use.