pmm_donor_selection: Score-based PMM donor selection

In VIM: Visualization and Imputation of Missing Values

Score-based PMM donor selection

Description

For each missing value, finds k nearest observed values based on model score (predicted value) distance, then aggregates using the chosen method.

Usage

pmm_donor_selection(
  y_obs,
  score_obs,
  score_miss,
  k = 1L,
  agg_method = "random"
)

Arguments

y_obs	Observed values of target variable
score_obs	Model scores for observed rows
score_miss	Model scores for missing rows
k	Number of nearest donors
agg_method	One of "mean", "median", "random", or a function

Value

Numeric vector of length(score_miss) with imputed values

VIM documentation built on Sept. 2, 2026, 5:07 p.m.