holisticai.bias.metrics.disparate_impact_regression#

holisticai.bias.metrics.disparate_impact_regression(group_a, group_b, y_pred, q=0.8)[source]#

Disparate Impact quantile (Regression version)

This function computes the ratio of success rates between group_a and group_b, where sucess means predicted score exceeds a given quantile (default = 0.8).

If q is a vector, this function returns a vector with the respective result for each given quantile in q.

Interpretation

A value of 1 is desired. Values below 1 are unfair towards group_a. Values above 1 are unfair towards group_b. The range (0.8,1.2) is considered acceptable.

Parameters

group_aarray-like: Group membership vector (binary)
group_barray-like: Group membership vector (binary)
y_predarray-like: Predictions vector (regression)
qfloat, array-like, optional: quantile of predictions considered, default=0.8

Returns

float: Disparate Impact (top %)

Notes

\(\frac{sr_a}{sr_b}\)

Examples

>>> import numpy as np
>>> from holisticai.bias.metrics import disparate_impact_regression
>>> group_a = np.array([1, 1, 1, 1, 0, 0, 0, 0, 0, 0])
>>> group_b = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1])
>>> y_pred = np.array([0.8, 0.9, 0.2, 0.1, 0.7, 0.9, 0.8, 0.6, 0.3, 0.5])
>>> disparate_impact_regression(group_a, group_b, y_pred, q=0.7)
1.5