DixonQTestDetector<T>
Models & Types · AiDotNet.AnomalyDetection.Statistical
Detects anomalies using Dixon’s Q Test for small datasets.
Inheritance: ModelBase<T, Matrix<T>, Vector<T>> → AnomalyDetectorBase<T> → DixonQTestDetector<T>
Implements: IGeneratedParameterRegistrar<T>
For Beginners
Dixon’s Q Test is designed for detecting a single outlier in small datasets (typically n < 25). It compares the gap between the suspect value and its nearest neighbor to the range of the entire dataset.
How It Works
The test statistic is: Q = gap / range where gap = |suspect value - nearest value| and range = max - min.
When to use:
- Small datasets (n < 25, ideally 3-10 samples)
- You suspect exactly one outlier
- Data is approximately normally distributed
Industry Standard Defaults:
- Alpha (significance level): 0.05 (5%)
Reference: Dixon, W. J. (1950). “Analysis of Extreme Values.” Annals of Mathematical Statistics.
Constructors
| Constructor | Summary |
|---|---|
DixonQTestDetector(double alpha, double contamination, int randomSeed) |
Creates a new Dixon’s Q Test anomaly detector. |
Properties
| Property | Summary |
|---|---|
double Alpha { get; } |
Gets the significance level (alpha) for the test. |
Methods
| Method | Summary |
|---|---|
Vector<T> Predict(Matrix<T> X) |
|
Vector<T> ScoreAnomalies(Matrix<T> X) |
|
double GetQCritical(int n) |
Gets the critical Q value for Dixon’s test based on sample size and alpha level. |
void Fit(Matrix<T> X) |