Can you please explain Kolmogorov-Smirnov & Anderson-Darling goodness of fit. In both the case, they are measuring the distance between the empirical distribution function & the CDF of the reference distribution. So, how are they different?
Hi Katherine, Sorry I am not unable to understand the concept of the same. It would be great if you can explain the same.
Please avoid addressing your questions directly to tutors. This forum is a place where students can ask help from each other, and so improve their own understanding. By asking us directly, it discourages others from answering. The Anderson-Darling statistic places more weight on differences in the tails of the distribution. Hence, the A-D test is more sensitive to deviations in the tails while the K-S test is more sensitive to deviations in the centre of the distribution.