dev.nm.stat.evt.evd.univariate.fitting.pot

Class PeaksOverThresholdOnClusters

java.lang.Object

dev.nm.stat.evt.evd.univariate.fitting.pot.PeaksOverThresholdOnClusters

All Implemented Interfaces:

MaximumLikelihoodFitting

public class PeaksOverThresholdOnClusters
extends Object
implements MaximumLikelihoodFitting

Similar to POT, but only use the peak observations in clusters for the parametric estimation. Peaks are defined as the maxima in the clusters found by ClusterAnalyzer.

The R equivalent functions are evd::fpot with cmax=TRUE.

See Also:

"Enrique Castillo, Ali S. Hadi, N. Balakrishnan, Jose M. Sarabia, "Extreme Value and Related Models with Applications in Engineering and Science," P.268"

Constructor Summary

Constructors

Constructor and Description
PeaksOverThresholdOnClusters(ClusterAnalyzer clusterAnalyzer) Create an instance with a ClusterAnalyzer which is used to find clusters from observations.

Method Summary

Modifier and Type	Method and Description
EstimateByLogLikelihood	fit(double[] observations) Fit the model with the given observations.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

PeaksOverThresholdOnClusters

public PeaksOverThresholdOnClusters(ClusterAnalyzer clusterAnalyzer)

Create an instance with a ClusterAnalyzer which is used to find clusters from observations.

Parameters:

clusterAnalyzer - the cluster analyzer

Method Detail

fit

public EstimateByLogLikelihood fit(double[] observations)
                            throws Exception

Description copied from interface: MaximumLikelihoodFitting

Fit the model with the given observations.

Specified by:

fit in interface MaximumLikelihoodFitting

Parameters:

observations - the observations for fitting

Returns:

the fitted parameters of the model

Throws:

Exception - when an error has occurred during fitting process