Class ACERByCounting
- java.lang.Object
-
- dev.nm.stat.evt.evd.univariate.fitting.acer.empirical.ACERByCounting
-
public class ACERByCounting extends Object
Estimate epsilons by counting conditional exceedances from the observations. The R equivalent function isacer::acer.estimate_acer
.
-
-
Constructor Summary
Constructors Constructor Description ACERByCounting(double[] barrierLevels, int kStepMemory)
Create an instance for estimating epsilon for each of the given barrier levels.
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description double[][]
estimateForMultiPeriods(double[][] observations)
Estimate for multiple periods.double[]
estimateForOnePeriod(double[] observations)
Estimate for a single period.
-
-
-
Method Detail
-
estimateForMultiPeriods
public double[][] estimateForMultiPeriods(double[][] observations)
Estimate for multiple periods. This method is suitable for both stationary and non-stationary time series. Estimates for multiple periods can then be used to compute the sample mean, standard deviation, and confidence interval.- Parameters:
observations
- data for multiple periods (one row for one period)- Returns:
- estimated ACER values, each row contains estimates for one period
-
estimateForOnePeriod
public double[] estimateForOnePeriod(double[] observations)
Estimate for a single period.- Parameters:
observations
- data for one period- Returns:
- estimated ACER values, the i-th element = \(\hat{\epsilon_k}(\eta_i)\)
-
-