one_boot_scrnp: Internal function used to perform one bootstrap sample. The...

In nlpred: Estimators of Non-Linear Cross-Validated Risks Optimized for Small Samples

Description

Internal function used to perform one bootstrap sample. The function trys to fit learner on a bootstrap sample. If for some reason (e.g., the bootstrap sample contains no observations with Y = 1) the learner fails, then the function returns NA. These NAs are ignored later when computing the bootstrap corrected estimate.

Usage

one_boot_scrnp(Y, X, n, correct632, learner, sens)

Arguments

Y	A numeric binary outcome
X	A data.frame of variables for prediction.
n	Number of observations
correct632	A boolean indicating whether to use the .632 correction.
learner	A wrapper that implements the desired method for building a prediction algorithm. See ?glm_wrapper or read the package vignette for more information on formatting learners.
sens	The sensitivity constraint to use.

Value

If learner executes successfully, a numeric estimate of AUC on this bootstrap sample. Otherwise the function returns NA.

nlpred documentation built on Feb. 24, 2020, 1:11 a.m.