Skip to contents

Assess Cluster Stability via Bootstrap

Source: R/clustering.R
assess_cluster_stability.Rd

Evaluates the stability of cluster assignments for a SINGLE selected model using bootstrap resampling. For each bootstrap sample, the clustering is re-run with the same model specification and assignments are compared to the original using the Adjusted Rand Index.

Note: This function assesses one model at a time. If you tested multiple models, use the model_name parameter to specify which model to assess, or leave NULL to use the best model by BIC.

Usage

assess_cluster_stability(
  results,
  model_name = NULL,
  n_boot = 100,
  verbose = TRUE,
  seed = NULL
)

Arguments

results

Object from clustering()

model_name

Model to assess (single model). If NULL, uses best model by BIC.

n_boot

Number of bootstrap iterations. Defaults to 100.

verbose

Print progress. Defaults to TRUE.

seed

Random seed for reproducibility. Defaults to NULL.

Value

A list of class "cluster_stability" containing:

  • overall_stability: Mean Adjusted Rand Index across bootstraps

  • stability_sd: Standard deviation of ARI

  • bootstrap_ari: Vector of ARI values for each bootstrap

  • observation_stability: Proportion of times each observation was assigned to the same cluster as in the original

  • cluster_stability: Stability score for each cluster

  • interpretation: Text interpretation of stability

Examples

if (FALSE) { # \dontrun{
results <- clustering(data, vars, n_clusters = 3)

# Assess stability of best model (default)
stability <- assess_cluster_stability(results, n_boot = 100)

# Assess a specific model
stability <- assess_cluster_stability(results, model_name = "VVV", n_boot = 50)

# View results
print(stability)
stability$overall_stability
stability$interpretation

# Plot observation stability
hist(stability$observation_stability, main = "Observation Stability")
} # }