Learn R Programming

mev (version 2.1)

tstab.lthill: Threshold stability plots for left-truncated Hill estimators

Description

Given a vector of exceedances and some potential choices of \(k\) for the threshold, compute the left-truncated Hill estimators for each value of k and use these to compute the variance and slope of the estimator

Usage

tstab.lthill(xdat, k, which = c("lthill", "var", "slope"), log = TRUE, ...)

Value

an invisible list with lthill, order statistics, the log variance and the log scale.

Arguments

xdat

[numeric] vector of positive observations

k

[integer] number of order statistics for the threshold

which

[string] the type of plot, showing the left-truncated Hill plot on the log, the log of the variance of the estimator, or the log slope

log

[logical] if TRUE (default), shows the Hill plot on the log-scale

...

additional parameters for color, etc. to be passed to plot

References

Bladt, M., Albrecher, H. & Beirlant, J. (2020) Threshold selection and trimming in extremes. Extremes, 23, 629-665 . tools:::Rd_expr_doi("10.1007/s10687-020-00385-0")

Examples

Run this code
xdat <- 10/(1 - runif(n = 1000)) - 10
tstab.lthill(xdat = xdat, k = c(50,100,200))

Run the code above in your browser using DataLab