Anderson-Darling test for normality

Description

Performs the Anderson-Darling goodness-of-fit test for the hypothesis of normality. The null hypothesis is that the sample comes from a normal distribution.

Hypothesis of Normality The hypothesis of normality refers to the null hypothesis that the data comes from a normal distribution. In the implementation, the statistic is computed from the sample passed to execute_statistic.

Test Statistic The statistic is based on ordered standardized observations with the log CDF and log survival function of the normal distribution.

Usage

from pysatl_criterion.statistics.goodness_of_fit import (
    AndersonDarlingNormalityGofStatistic,
)


test_statistic = AndersonDarlingNormalityGofStatistic()
statistic_result = test_statistic.execute_statistic([-1.21, -0.83, -0.52, -0.31, -0.08, 0.14, 0.29, 0.47, 0.68, 0.91, 1.16, 1.43])
print(statistic_result)

Arguments

mean - reference normal mean. Default value is 0.

var - reference normal variance. Default value is 1.

rvs - array-like sample data passed to execute_statistic.

Details

The implementation evaluates the Anderson-Darling statistic for the supplied observations. Large or small values should be interpreted according to the statistic alternative used by the class implementation.

References

The statistic follows the implementation in pysatl_criterion.statistics.goodness_of_fit.normal.

Author(s)

Alexey Mironov

Examples

from pysatl_criterion.statistics.goodness_of_fit import (
    AndersonDarlingNormalityGofStatistic,
)


test_statistic = AndersonDarlingNormalityGofStatistic()
statistic_result = test_statistic.execute_statistic([-1.21, -0.83, -0.52, -0.31, -0.08, 0.14, 0.29, 0.47, 0.68, 0.91, 1.16, 1.43])
print(statistic_result)