bayesplain.frequentist.two_means

bayesplain.frequentist.two_means(x, y, equal_var=False, level=0.95)[source]

Two-sample t-test and confidence interval on the difference.

Defaults to Welch’s unequal-variance test, which is the honest default and the one that matches the Behrens-Fisher posterior on the Bayesian side.

Parameters:
x : array_like

The two samples. The difference is oriented as y - x.

y : array_like

The two samples. The difference is oriented as y - x.

equal_var : bool, default False

Assume equal variances and pool them (Student’s t) rather than using Welch’s correction.

level : float, default 0.95

Confidence level.

Returns:

The conventional result, oriented as group 2 minus group 1.

Return type:

FrequentistTwin