Kolmogorov-Smirnov Test

The two-sample Kolmogorov-Smirnov test is used to detect if two samples come from the same underlying distribution. More precisely, this non-parametric test calculates a distance d between the empirical distribution functions of the two samples.

The corresponding p-value can be computed exactly if there are no ties (duplicate values) present in the samples and the product of the two sample sizes is less than 10000. Otherwise, the p-value has to be approximated with the given options.

With the given significance level α and the calculated p-value the null hypothesis H0 (the two samples come from a common distribution) will either be rejected or not.

Please refer also to the Wikipedia description of the Kolmogorov-Smirnov Test.

Options

Significance level α
Significance level at which the null hypothesis can be rejected, 0 < α < 1
First test column
The first sample (column) included in the test
Second test column
The second sample (column) included in the test

Advanced Settings

Missing values strategy
How missing values in the samples are going to be handled:
  • REMOVED - missing values are ignored
  • FAILED - missing values result in node failure
Exact p-value
If the exact p-value should be computed. This requires:
  • No ties present in both samples
  • Product of the two sample sizes < 10000
Cauchy criterion
Stopping criterion for the approximation of the p-value when successive partial sums are within tolerance
Max number of iterations
Stopping criterion for the approximation of the p-value when the number of iteration is reached

Input Ports

Icon
The table from which to test two samples

Output Ports

Icon
Kolmogorov-Smirnov test evaluation

Views

This node has no views

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.