statsmodels.stats.diagnostic.anderson_statistic¶

statsmodels.stats.diagnostic.anderson_statistic(x, dist='norm', fit=True, params=(), axis=0)[source]¶

Calculate the Anderson-Darling a2 statistic.

Parameters:¶

x : array_like¶

The data to test.

dist : {'norm', callable}¶

The assumed distribution under the null of test statistic.

fit : bool¶

If True, then the distribution parameters are estimated. Currently only for 1d data x, except in case dist=’norm’.

params : tuple¶

The optional distribution parameters if fit is False.

axis : int¶

If dist is ‘norm’ or fit is False, then data can be an n-dimensional and axis specifies the axis of a variable.

Returns:¶

The Anderson-Darling statistic.

Return type:¶

{float, ndarray}