WilcoxonSignedRankTest<T>
Models & Types · AiDotNet.Evaluation.Statistics
Wilcoxon signed-rank test: non-parametric paired comparison test.
Implements: IPairedTest<T>
For Beginners
This is the non-parametric alternative to the paired t-test. Use it when:
- Your differences are not normally distributed
- You have ordinal data
- Your sample size is small and you can’t verify normality
- You want a more robust test (less sensitive to outliers)
How It Works
Common use in ML: Comparing cross-validation scores of two models when you’re not sure the performance differences are normally distributed.
Constructors
| Constructor | Summary |
|---|---|
WilcoxonSignedRankTest() |
Properties
| Property | Summary |
|---|---|
string Description { get; } |
|
string Name { get; } |
Methods
| Method | Summary |
|---|---|
StatisticalTestResult<T> Test(T[] sample1, T[] sample2, double alpha) |