extremal: Estimate extremal index using 'intervals' method

In evgam: Generalised Additive Extreme Value Models

Estimate extremal index using ‘intervals’ method

Description

Estimate extremal index using ‘intervals’ method

Usage

extremal(x, y = NULL)

Arguments

x	a logical vector or list of logical vectors
y	an integer vector the same length as x; see Details

Details

Intervals estimator of extremal index based on Ferro and Segers (2003)'s moment-based estimator.

If x is supplied and y is not, x is assumed to identify consecutive threshold exceedances. If x is supplied as a list, each list element is assumed to comprise identifiers of consecutive exceedances. If y is supplied, x must be a logical vector, and y gives positions of x in its original with-missing-values vector: so y identifies consecutive x.

Value

A scalar estimate of the extremal index

References

Ferro, C. A., & Segers, J. (2003). Inference for clusters of extreme values. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 65(2), 545-556.

Examples

n <- 1e2
x <- runif(n)
extremal(x > .9)

y <- sort(sample(n, n - 5))
x2 <- x[y]
extremal(x2 > .9, y)

evgam documentation built on June 28, 2022, 5:07 p.m.