Skip to contents

Identify Variable Clusters for Policy Tree Analysis

Source: R/margot_policy_tree_diagnostics.R
margot_identify_variable_clusters.Rd

Groups variables into clusters based on their correlations, helping to understand which variables might substitute for each other in policy tree splits.

Usage

margot_identify_variable_clusters(
  correlation_assessment,
  min_cluster_size = 2,
  label_mapping = NULL
)

Arguments

correlation_assessment

Output from margot_assess_variable_correlation()

min_cluster_size

Minimum number of variables to form a cluster (default 2)

label_mapping

Optional named list mapping variable names to labels. If NULL and correlation_assessment contains label_mapping, uses that; otherwise uses automatic transformation

Value

Data frame with cluster assignments and representative variables (using labels)