PredictiveCheckTest.extreme#

PredictiveCheckTest.extreme(*, level=_EXTREME_PVALUE)[source]#

Labels statistic[variable] with a tail probability outside (level, 1 - level).

Parameters:

level (float) – The tail size on each side; the default marks a cell whose p-value is below 0.05 or above 0.95.

Return type:

tuple[str, ...]

Returns:

The labels in row-major order over statistics and variables; empty when nothing sits in a tail. A nan p-value is never flagged.

Raises:

SpecificationError – If the level is not inside (0, 0.5).

Example

>>> import numpy as np
>>> rng = np.random.default_rng(3)
>>> y = 2.0 + rng.standard_normal((100, 1))
>>> check = replication_check(y, rng.standard_normal((200, 100, 1)))
>>> check.extreme()
('mean[y]',)