Kolmogorov-Smirnov Test
The K-S test is particularly useful when sample sizes are small and when no parameters have been estimated from the data.
The basic technique of this test is to determine the difference between the cdf, Sn(x) of the sample (collected) data and the cdf, F(x) assumed for the population from which the samples were taken.
When a distributional assumption is rejected, another distribution is tried. When all else fails, the empirical distribution may be used.