piRF (version 0.1.0)

GhosalBoostRF: Implements RF prediction interval method in Ghosal, Hooker 2018. Helper function.

Description

This function implements variant one and two of the prediction interval methods in Ghosal, Hooker 2018. Used inside rfint().

Usage

GhosalBoostRF(
  formula = NULL,
  train_data = NULL,
  pred_data = NULL,
  num_trees = NULL,
  min_node_size = NULL,
  m_try = NULL,
  keep_inbag = TRUE,
  intervals = FALSE,
  alpha = NULL,
  prop = NULL,
  variant = 1,
  num_threads = NULL,
  num_stages = NULL,
  interval_type = NULL
)

Arguments

formula

Object of class formula or character describing the model to fit. Interaction terms supported only for numerical variables.

train_data

Training data of class data.frame, matrix, dgCMatrix (Matrix) or gwaa.data (GenABEL). Matches ranger() requirements.

pred_data

Test data of class data.frame, matrix, dgCMatrix (Matrix) or gwaa.data (GenABEL). Utilizes ranger::predict() to get prediction intervals for test data.

num_trees

Number of trees.

min_node_size

Minimum number of observations before split at a node.

m_try

Number of variables to randomly select from at each split.

keep_inbag

Saves matrix of observations and which tree(s) they occur in. Required to be true to generate variance estimates for Ghosal, Hooker 2018 method.

intervals

Generate prediction intervals or not. Defaults to FALSE.

alpha

Significance level for prediction intervals.

prop

Proportion of training data to sample for each tree. Currently variant 2 not implemented.

variant

Choose which variant to use. Currently variant 2 not implemented.

num_threads

The number of threads to use in parallel. Default is the current number of cores.

num_stages

Number of boosting stages. Functional for >= 2; variance estimates need adjustment for variant 2.

interval_type

Type of prediction interval to generate. Options are method = c("two-sided", "lower", "upper"). Default is method = "two-sided".