Generate Meinshausen hierarchical clustering for variables

Creates a hierarchical clustering structure suitable for Meinshausen testing when applied to variable selection problems. This is typically used when the goal is to test groups of correlated variables hierarchically.

Usage

generate_meinshausen_hierarchy(
  correlation_matrix,
  method = "complete",
  min_cluster_size = 2
)

Arguments

correlation_matrix

Correlation matrix of variables

method

Clustering method ("complete", "average", "single")

min_cluster_size

Minimum size for clusters to be tested

Value

List with clustering structure compatible with node_tracker format