fit.shape: Shape parameter estimates

In lbelzile/mev: Modelling of Extreme Values

Shape parameter estimates

Description

Wrapper to estimate the tail index or shape parameter of an extreme value distribution. Each function has similar sets of arguments, a vector or scalar number of order statistics k and a vector of positive observations xdat. The method argument allows users to choose between different indicators, including the Hill estimator (hill, for positive observations and shape only), the moment estimator of Dekkers and de Haan (mom or dekkers), the Beirlant, Vynckier and Teugels generalized quantile estimator (bvt or genquant), the Pickands estimator (pickands), the extreme U-statistics estimator of Oorschot, Segers and Zhou (osz, or pickandsxu).

Usage

fit.shape(
  xdat,
  k,
  method = c("hill", "pickandsxu", "osz", "vries", "mom", "dekkers", "bvt", "genquant",
    "pickands"),
  ...
)

Arguments

xdat	vector of positive observations of length n
k	number of largest order statistics
method	estimation method.
...	additional parameters passed to functions

Value

a data frame with the number of order statistics k and the shape parameter estimate shape, or a single numeric value if k is a scalar.

lbelzile/mev documentation built on June 14, 2025, 6:40 p.m.