Example
The following example illustrates the class KolmogorovTwoSample
routine with two randomly generated samples from a uniform(0,1) distribution. Since the two theoretical distributions are identical, we would not expect to reject the null hypothesis.
import com.imsl.stat.*;
public class KolmogorovTwoSampleEx1 {
static public void main(String arg[]) {
double x[] = new double[100];
double y[] = new double[60];
Random random = new Random(123457);
random.setMultiplier(16807);
for (int i = 0; i < x.length; i++) {
x[i] = random.nextFloat();
}
for (int i = 0; i < y.length; i++) {
y[i] = random.nextFloat();
}
KolmogorovTwoSample k2s = new KolmogorovTwoSample(x, y);
System.out.println("D = "+ k2s.getTestStatistic());
System.out.println("D+ = " + k2s.getMaximumDifference());
System.out.println("D- = " + k2s.getMinimumDifference());
System.out.println("Z = " + k2s.getZ());
System.out.println("Prob greater D one sided = " +
k2s.getOneSidedPValue());
System.out.println("Prob greater D two sided = " +
k2s.getTwoSidedPValue());
System.out.println("Missing X = " + k2s.getNumberMissingX());
System.out.println("Missing Y = " + k2s.getNumberMissingY());
}
}
Output
D = 0.18
D+ = 0.18
D- = 0.010000000000000009
Z = 1.1022703842524302
Prob greater D one sided = 0.07201060734868497
Prob greater D two sided = 0.14402121469736995
Missing X = 0
Missing Y = 0
Link to Java source.