## Anderson-Darling StatisticCompute the probability associated with the Anderson-Darling statistic: \(AD_{Statistic} = - n - \frac{1}{n} \cdot \sum_{i=0}^{n-1}{(2 \cdot i + 1 ) \cdot ln(f_{i} \cdot (1 - f_{n-i-1}) )} \\ when \: 0 < f_{i} \leq f_{i+1} < 1, \, for \; 0 \leq i \leq n-1 \)(f = cummulative distribution function of the distribution function being tested). Computation limitations:Sample size, n; 2 ≤ n ≤ 1000, not necessary an integer.Calculated value of the Anderson-Darling statistic, on the sample of size n, with at least five significant digits, AD; 0.1 ≤ AD ≤ 10.Return value:Probability to be observed a better agreement between the observed sample and the hypothetical distribution being tested.Is obtained with four significant digits.Limitation: 1.0E-11 ≤ min(p,1-p).Ref:A paper describing the procedure will be prepared.Calculation uses 30 coeficients obtained from a high resolution Monte-Carlo experiment.Compute for: | |