The SDL Component Suite is an industry leading collection of components supporting scientific and engineering computing. Please visit the SDL Web site for more information....



KolmogSmir1SampleTestStat


Unit:SDL_statis
Class:none
Declaration:function KolmogSmir1SampleTestStat (SampleSize: integer; alpha: TSignifLevel): double;

A frequently encountered problem in statistics is to test whether a given number of samples belong to a predefined probability distribution. The function KolmogSmir1SampleTestStat returns an approximation of the critical values for the Kolmogorov-Smirnov test (KS test). The parameter SampleSize specifies the number of samples used for the test, the parameter alpha specifies the level of uncertainty.

Note that the test statistic is calculated by finding the maximum absolute difference between the cumulative frequency distribution of the samples and the predefined distribution.

Hint: A more accurate (but slower) routine to calculate the quantiles of the KS test is the function KSQuantile.



Last Update: 2012-Oct-20