scikit_posthocs.global_f_test

scikit_posthocs.global_f_test(p_vals: Union[List[T], numpy.ndarray], stat: bool = False) → Union[float, Tuple[float, float]]

Fisher’s combination test for global null hypothesis.

Computes the combined p value using chi-squared distribution and T statistic: -2 * sum(log(x)) [1].

Parameters:
  • p_vals (Union[List, ndarray]) – An array or a list of p values.
  • stat (bool) – Defines if statistic should be returned.
Returns:

  • p_value (float) – Global p value.
  • t_stat (float) – Statistic.

References

[1]Fisher RA. Statistical methods for research workers, London: Oliver and Boyd, 1932.

Examples

>>> x = [0.04, 0.03, 0.98, 0.01, 0.43, 0.99, 1.0, 0.002]
>>> sp.global_f_test(x)