hidimstat.ensemble_clustered_inference_pvalue#
- hidimstat.ensemble_clustered_inference_pvalue(n_samples, group, list_ward, list_beta_hat, list_theta_hat, list_precision_diag, aggregate_method=<function adaptive_quantile_aggregation>, n_jobs=None, verbose=0, **kwargs)[source]#
Compute and aggregate p-values across multiple bootstrap iterations using an aggregation method.
This function performs statistical inference on each bootstrap sample and combines the results using a specified aggregation method to obtain robust estimates. The implementation follows the methodology in [1].
- Parameters:
- n_samplesint
Number of samples in the dataset
- groupbool
If True, uses group lasso p-values for multivariate outcomes
- list_wardlist of AgglomerativeClustering
List of fitted clustering objects from bootstraps
- list_beta_hatlist of ndarray
List of estimated coefficients at cluster level from each bootstrap
- list_theta_hatlist of ndarray
List of estimated precision matrices from each bootstrap
- list_precision_diaglist of ndarray
List of diagonal elements of covariance matrices from each bootstrap
- aggregate_methodcallable, default=adaptive_quantile_aggregation
Function to aggregate results across bootstraps. Must accept a 2D array and return a 1D array of aggregated values.
- n_jobsint or None, optional (default=None)
Number of parallel jobs. None means using all processors.
- verboseint, default=0
Verbosity level for computation progress
- **kwargsdict
Additional arguments passed to p-value computation functions
- Returns:
- beta_hatndarray, shape (n_features,) or (n_features, n_times)
Averaged coefficients across bootstraps
- pvalndarray, shape (n_features,)
Aggregated p-values for each feature
- pval_corrndarray, shape (n_features,)
Aggregated multiple testing corrected p-values
- one_minus_pvalndarray, shape (n_features,)
Aggregated 1-p values for numerical stability
- one_minus_pval_corrndarray, shape (n_features,)
Aggregated 1-corrected p values for numerical stability
References
