mev (version 2.1)

tstab.egp: Threshold stability plots for extended generalized Pareto models

Description

Threshold stability plots for extended generalized Pareto models

Usage

tstab.egp(
  xdat,
  thresh,
  model = c("pt-beta", "pt-gamma", "pt-power", "gj-tnorm", "gj-beta", "exptilt",
    "logist"),
  param = c("shape", "kappa"),
  type = c("wald", "lrt"),
  transform = FALSE,
  level = 0.95,
  plot = TRUE,
  ...
)

Value

an invisible list object of class mev_egp_tstab with elements

• kappa: matrix of parameter estimates and confidence intervals for \(\kappa\), if specified in param

• shape: matrix of parameter estimates and confidence intervals for the shape parameter \(\xi\), if specified in param

• thresh: vector of threshold candidates

• level: level of the confidence intervals

• model: string giving the EGP model family

• type: type of confidence interval

Arguments

xdat

vector of observations, greater than the threshold

thresh

threshold value

model

a string indicating which extended family to fit

param

[string] parameter, either shape or additional parameter kappa

type

[string] confidence interval type, either wald or profile.

transform

logical; if TRUE and type="wald", intervals for kappa are computed on the log-scale and back-transformed.

level

[double] confidence interval level, default to 0.95.

plot

[logical] if TRUE (default), return a threshold stability plot

...

additional arguments for the plot function, currently ignored.

Examples

xdat <- rgp(n = 1000)
tstab.egp(
 xdat = xdat,
 thresh = c(0, quantile(xdat, 0.5)),
 model = "gj-tnorm",
 param = "kappa",
 transform = TRUE)