public class KruskalWallis extends HypothesisTest
Since it is a non-parametric method, the Kruskal–Wallis test does not assume a normal population, unlike the analogous one-way analysis of variance. However, the test does assume an identically-shaped and scaled distribution for each group, except for any difference in medians.
TODO: correction for ties
The R equivalent function is kruskal.test.
k, N, pValue, testStatistics| Constructor and Description |
|---|
KruskalWallis(double[]... samples)
Construct a Kruskal-Wallis test for the equality of median of groups.
|
| Modifier and Type | Method and Description |
|---|---|
java.lang.String |
getAlternativeHypothesis()
Get a description of the alternative hypothesis.
|
java.lang.String |
getNullHypothesis()
Get a description of the null hypothesis.
|
isNullRejected, oneSidedPvalue, pValue, statisticspublic KruskalWallis(double[]... samples)
samples - samplespublic java.lang.String getNullHypothesis()
HypothesisTestgetNullHypothesis in class HypothesisTestpublic java.lang.String getAlternativeHypothesis()
HypothesisTestgetAlternativeHypothesis in class HypothesisTest