process_fold_mean_diff: Process one cross-validation fold for mean difference testing

In HMC: High-Dimensional Mean Comparison with Projection and Cross-Fitting

Process one cross-validation fold for mean difference testing

Description

Computes the test statistic, variance, and projection direction for one fold in a cross-validated comparison of two groups.

Usage

process_fold_mean_diff(
  fold_index,
  control,
  treatment,
  control_split_index,
  tr_split_index,
  pca_method,
  classifier_method,
  lambda_type,
  group,
  verbose
)

Arguments

fold_index	Integer index of the current fold.
control	Matrix or data frame for the control group (rows = samples, columns = features).
treatment	Matrix or data frame for the treatment group (rows = samples, columns = features).
control_split_index	A list of row indices for each fold of the control group.
tr_split_index	A list of row indices for each fold of the treatment group.
pca_method	Character. PCA method to use. Options are "dense_pca" or "sparse_pca".
classifier_method	Character. Classifier method. Options are "lasso" or "group_lasso".
lambda_type	Character. Lambda selection method. Options are "lambda.min" or "lambda.1se".
group	Optional grouping vector for group lasso.
verbose	Logical. Whether to print progress messages.

Value

A list containing the test statistic, its variance, scores for each group, the projection direction, and intermediate quantities.

HMC documentation built on June 8, 2025, 10:32 a.m.