public class Outliers extends Object
A class of static methods to produce estimations of the scale of variation within a Dataset. The available estimators are:
Croux, C. and P. J. Rousseeuw, "Time-efficient algorithms for two highly robust estimators of scale", Computational Statistics, Volume 1, eds. Y. Dodge and J.Whittaker, Physica-Verlag, Heidelberg, pp411--428 (1992).
Constructor and Description |
---|
Outliers() |
Modifier and Type | Method and Description |
---|---|
static double |
highMed(Dataset data)
Returns the himed
|
static double |
lowMed(Dataset data)
Returns the lomed
|
static double[] |
medianAbsoluteDeviation(Dataset data)
Returns the Median Absolute Deviation (MAD) and the median.
|
static double |
medianOFTwoPrimitiveArrays(double[] a,
double[] b)
Calculates the overall median of two double arrays
|
static double |
snFast(Dataset data)
Returns the Sn estimator of Croux and Rousseeuw.
|
static double |
snNaive(Dataset data)
Returns the Sn estimator of Croux and Rousseeuw.
|
public Outliers()
public static double[] medianAbsoluteDeviation(Dataset data)
data
- The data for which the median and the MAD are to be calculatedpublic static double snNaive(Dataset data)
This is the simple O(n²) version of the calculation algorithm.
data
- The data for which the estimator is to be calculated.public static double snFast(Dataset data)
This is the complex O(nlog n) version of the calculation algorithm.
data
- The data for which the estimator is to be calculated.public static double lowMed(Dataset data)
Returns the lomed (low median) of a sorted Dataset.
data
- A sorted Dataset for which the low median is to be calculated.public static double highMed(Dataset data)
Returns the himed (high median) of a sorted Dataset.
data
- A sorted Dataset for which the low median is to be calculated.public static double medianOFTwoPrimitiveArrays(double[] a, double[] b)
a
- b
- the two arrays for which the overall median is desired.Copyright © 2014–2019 Eclipse Foundation. All rights reserved.