RunsTest<T> Class

public sealed class RunsTest<T> : OneSampleTest<T>

Public NotInheritable Class RunsTest(Of T)
	Inherits OneSampleTest(Of T)

generic<typename T>
public ref class RunsTest sealed : public OneSampleTest<T>

[<SealedAttribute>]
type RunsTest<'T> = 
    class
        inherit OneSampleTest<'T>
    end

Inheritance

Object  →  HypothesisTest  →  OneSampleTest<T>  →  RunsTest<T>

Type Parameters

T

The element type of the sample.

Use the RunsTest<T> class to test whether the elements of a sample are randomly distributed. Specifically, the test counts the number of sequences ("runs") of values that are larger or smaller than a specific value, and compares it to the expected number of runs for a random sequence.

For example, the residuals of a regression calculation should be randomly distributed around zero. If the fitted line or curve is not a good model for the data, then the residuals will tend to show long runs of positive or negative values. The runs test can detect this pattern.

The distribution of the number of runs is approximately normal. For small sample sizes (less than 50) a continuity correction is applied.

The runs test is sometimes called the Wald-Wolfowitz test.

Reference