Package elki.math.statistics.distribution.estimator

Interface MADDistributionEstimator<D extends Distribution>

Type Parameters:

D - Distribution estimated.

All Superinterfaces:

DistributionEstimator<D>

All Known Implementing Classes:

CauchyMADEstimator, ExponentialMADEstimator, ExponentialMedianEstimator, GumbelMADEstimator, LaplaceMADEstimator, LogisticMADEstimator, NormalMADEstimator, RayleighMADEstimator, UniformMADEstimator

public interface MADDistributionEstimator<D extends Distribution>
extends DistributionEstimator<D>

Distribuition estimators that use the method of moments (MOM), i.e. that only need the statistical moments of a data set.

Since:

0.6.0

Author:

Erich Schubert

Method Summary

Modifier and Type	Method	Description
static double	computeMAD(double[] x, int len, double median)	Compute the median absolute deviation from median.
static double	computeMAD(double[] data, int len, double median, double[] scratch)	Compute the median absolute deviation from median.
default <A> D	estimate(A data, NumberArrayAdapter<?,A> adapter)	General form of the parameter estimation
D	estimateFromMedianMAD(double median, double mad)	General form of the parameter estimation

Methods inherited from interface elki.math.statistics.distribution.estimator.DistributionEstimator

estimate, getDistributionClass

Method Detail

estimateFromMedianMAD

D estimateFromMedianMAD(double median,
                        double mad)

General form of the parameter estimation

Parameters:

median - Median value

mad - Median absolute deviation from median

Returns:

Estimated distribution

estimate

default <A> D estimate(A data,
                       NumberArrayAdapter<?,A> adapter)

Description copied from interface: DistributionEstimator

General form of the parameter estimation

Specified by:

estimate in interface DistributionEstimator<D extends Distribution>

Parameters:

data - Data set

adapter - Number array adapter

Returns:

Estimated distribution

computeMAD

static double computeMAD(double[] data,
                         int len,
                         double median,
                         double[] scratch)

Compute the median absolute deviation from median.

Parameters:

data - Input data

len - Length of input data to use

median - Median of input data

scratch - Scratch space, must be at least length len

Returns:

MAD

computeMAD

static double computeMAD(double[] x,
                         int len,
                         double median)

Compute the median absolute deviation from median.

Parameters:

x - Input data will be modified

len - Length where x is valid

median - Median value.

Returns:

Median absolute deviation from median.