be a random sample from a population with unknown
continuous CDF F(x) and assume an interest in testing the hypothesis that F(x)
corresponds to a known and completely specified distribution F0(x) against the
alternative that this is not true. The testing procedure proposed by Kolmogorov
(1933) is based on the supremum of the vertical distance between F0(x) and the
EDF based on the observed sample X. Smirnov (1939) proposed an extension of
the Kolmogorov test for comparing the distributions of two independent populations.
Statistics based on the vertical distance between F0(x) and the EDF are
called Kolmogorov-type statistics, while similar statistics based on the vertical distance
between two EDFs are called Smirnov-type statistics (Conover, 1999). The
Kolmogorov goodness-of-fit test presented in this paragraph is also called the onesample
Kolmogorov–Smirnov test. Formally the problem consists of testing the null
hypothesis