boot_MI: Bootstrap validation in Multiply Imputed datasets

In mwheymans/psfmi: Prediction Model Pooling, Selection and Performance Evaluation Across Multiply Imputed Datasets

Bootstrap validation in Multiply Imputed datasets

Description

boot_MI Bootstrapping followed by Multiple Imputation for internal validation. Called by function psfmi_perform.

Usage

boot_MI(
  pobj,
  data_orig,
  nboot = 10,
  nimp_mice,
  p.crit,
  direction,
  miceImp,
  ...
)

Arguments

pobj	An object of class pmods (pooled models), produced by a previous call to psfmi_lr.
data_orig	dataframe of original dataset that contains missing data.
nboot	The number of bootstrap resamples, default is 10.
nimp_mice	Numerical scalar. Number of multiple imputation runs.
p.crit	A numerical scalar. P-value selection criterium used for backward or forward selection during validation. When set at 1, validation is done without variable selection.
direction	The direction of predictor selection, "BW" is for backward selection and "FW" for forward selection.
miceImp	Wrapper function around the mice function.
...	Arguments as predictorMatrix, seed, maxit, etc that can be adjusted for the mice function.

Details

This function bootstraps from the incomplete dataset and applies MI in each bootstrap sample. The model that is selected by the psfmi_lr function is validated. When p.crit != 1, internal validation is conducted with variable selection. The performance measures in the multiply imputed bootstrap samples are tested in the original multiply imputed datasets (pooled) to determine the optimism.

Author(s)

Martijn Heymans, 2020

See Also

psfmi_perform

mwheymans/psfmi documentation built on June 30, 2023, 5:25 a.m.