public class ACERAnalysis extends Object
This algorithm works as follows:
Another wellknown estimation method is Peaks Over Threshold (POT). POT method assumes independence among extreme events, and therefore always requires declustering and dropping other nonpeak data. This is considered to be wasteful. On the other hand, ACER method accounts for Markovlike dependence (i.e., kstep memory) in time series (with k=1 as a special case for event independence). That is, a threshold exceedance is considered as an occurrence if the previous (k1) points are below the threshold. Experiments show that k=2 (i.e., conditional on one previous nonexceedance) is accurate enough for estimation for a wide range of data.
The R equivalent function is acer::acer.analysis
.
Modifier and Type  Class and Description 

static class 
ACERAnalysis.Result 
Constructor and Description 

ACERAnalysis()
Create an instance with the default values.

ACERAnalysis(int kStepMemory,
int nLevels,
double confidenceLevel,
boolean usePeaksOnly,
boolean weightedByPeakCount)
Create an instance with various options listed below.

Modifier and Type  Method and Description 

ACERAnalysis.Result 
run(double[][] observations,
double tailMarker)
Run the analysis with multiperiod observations.

ACERAnalysis.Result 
run(double[] observations,
double tailMarker)
Run the analysis with singleperiod observations.

public ACERAnalysis()
this(2, 300, 0.95, true, true);
public ACERAnalysis(int kStepMemory, int nLevels, double confidenceLevel, boolean usePeaksOnly, boolean weightedByPeakCount)
kStepMemory
 the value of k in the kstep memory modelnLevels
 the number of barrier levels to be used for estimationconfidenceLevel
 the confidence level for computing confidence intervalusePeaksOnly
 true
if use only peaks in the observations for estimationweightedByPeakCount
 true
if weight periods by the peak counts in the periodspublic ACERAnalysis.Result run(double[] observations, double tailMarker)
observations
 the observations (one row for each period, can has different length)tailMarker
 the appropriately chosen tail level \(\eta_1\)public ACERAnalysis.Result run(double[][] observations, double tailMarker)
observations
 the observations (one row for each period, can has different length)tailMarker
 the appropriately chosen tail level \(\eta_1\)Copyright © 20102018 Numerical Method Incorporation Limited. All Rights Reserved.