Anderson Darling Normality Test A Squared. For a specified data set and distribution the better the distribution fits the data the smaller this statistic will be. The statistic A was introduced by Anderson and Darling 1952 1950 and for Case 0 they gave the asymptotic distribution and tables of percentage points.
The Anderson-Darling Test measures the area between a fitted line based on the chosen distribution and a nonparametric step function based on the plot points. The p-value cell G14 is calculated as. The Anderson-Darling test Stephens 1974 is used to test if a sample of data came from a population with a specific distribution.
Here is the Anderson-Darling output for our data set.
The Anderson-Darling goodness-of-fit statistic AD-Value measures the area between the fitted line based on the normal distribution and the empirical distribution function which is based on the data points. Clearly rejecting Normality in a case like this is inappropriate. Large discrepancies between the EDF and the tested distribution will indicate a bad fit. So if you get an A-squared that is fairly large then you will get a small p-value and thus reject the null hypothesis.
