The null hypothesis of the Anderson–Darling test is that a dataset comes from a certain distribution; the reference distribution can be specified explicitely (one-sample test). K-sample Anderson–Darling tests are available for testing whether several samples are coming from a single population drawn from the distribution function which does not have to be specified.
Perform a one sample Anderson–Darling test of the null hypothesis that the data in vector
xcomes from the distribution
dagainst the alternative hypothesis that the sample is not drawn from
Perform an k-sample Anderson–Darling test of the null hypothesis that the data in vectors
xscomes from the same distribution against the alternative hypothesis that the samples comes from different distributions.
modifiedparamater enables a modified test calculation for samples whose observations do not all coincide.
- k-Sample Anderson-Darling Tests, F. W. Scholz and M. A. Stephens, Journal of the American Statistical Association, Vol. 82, No. 399. (Sep., 1987), pp. 918-924.