nispace.stats.effectsize.hedges
- nispace.stats.effectsize.hedges(a, b)[source]
Hedges’ g: small-sample bias-corrected Cohen’s d for independent groups.
Applies the exact correction factor 1 - 3/(4*dof - 1) to cohen(a, b).
- Parameters:
a (array_like, shape (n_obs_a, n_features) / (n_obs_b, n_features)) – Same convention as cohen.
b (array_like, shape (n_obs_a, n_features) / (n_obs_b, n_features)) – Same convention as cohen.
- Returns:
g
- Return type:
np.ndarray, shape (n_features,)
Notes
Uses plain cohen internally – NaN in a/b propagates. Not called anywhere in NiSpace’s internal pipeline: Y_transform=”hedges(a,b)” dispatches to the numba-jitted hedges_nan_fast (core/transform_y.py), not this function.
References