calibrate: calibrate()

In piRF: Prediction Intervals for Random Forests

Description

This function outputs a calibrated significance level based on coverage of prediction intervals generated using oob collections. Primarily for use in RoyRF().

Usage

calibrate(
  oob,
  alpha = 0.1,
  response_data,
  tolerance = 0.025,
  step_percent = 0.618,
  undercoverage = FALSE,
  method = "quantile",
  max_iter = 10
)

Arguments

oob	collection of oob predictions for training data (in list form).
alpha	nominal significance level. Defaults to 0.01.
response_data	response data of class data.frame. Must have names() attribute.
tolerance	tolerance allowed around nominal alpha. Default is 0.25.
step_percent	ratio absolute difference between empirical oob coverage and nominal coverage to adjust when calibrating. Defaults to 0.618.
undercoverage	Allow undercoverage. Defaults to TRUE. Not currently implemented.
method	Method to calibrate prediction intervals with. Defaults to "quantile"). Current only "quantile" implemented.
max_iter	Maximum number of iterations. Defaults to 10.

piRF documentation built on July 1, 2020, 7:51 p.m.