pint.utils.anderson_darling

pint.utils.anderson_darling(x: ndarray, mean: float = 0, variance: float = 1) → Tuple[float, float][source]

Anderson-Darling statistic with a known mean and variance for a normal distribution

Differs from scipy.stats.anderson() in that the mean and variance are specified, rather than fit.

Parameters:

  • x (np.ndarray) – Data

  • mean (float, optional) – Known mean of data

  • variance (float, optional) – Known variance of data

Returns:

  • float (test statistic)

  • float (p-value (lower means less likely to have come from a normal distribution))

References