ABSTRACT
There are many situations in which it is not possible for the statistician to make a rigid assumption about the shape of the population from which samples are being drawn. This limitation has led to the development of a group of alternate techniques known as nonparametric or distribution-free methods.
