GEV_shape_plot: Fitted GEV Shape as a Function of the Threshold

Description

Fit GEVs to block maxima and plot the fitted GPD shape as a function of the block size.

Usage

GEV_shape_plot(x, blocksize = tail(pretty(seq_len(length(x)/20), n = 64), -1),
               estimate.cov = TRUE, conf.level = 0.95,
               lines.args = list(lty = 2), xlab = "Block size",  ylab = NULL,
               xlab2 = "Number of blocks", plot = TRUE, ...)

Arguments

numeric vector of data.

blocksize

numeric vector of block sizes for which to fit a GEV to the block maxima.

estimate.cov

logical indicating whether confidence intervals are to be computed.

conf.level

confidence level of the confidence intervals if estimate.cov.

lines.args

list of arguments passed to the underlying lines() for drawing the confidence intervals.

xlab

x-axis label.

ylab

y-axis label (if NULL, a default is used).

xlab2

label of the secondary x-axis.

plot

logical indicating whether a plot is produced.

…

additional arguments passed to the underlying plot().

Value

Invisibly returns a list containing the block sizes considered, the corresponding block maxima and the fitted GEV distribution objects as returned by the underlying fit_GEV_MLE().

Details

Such plots can be used in the block maxima method for determining the optimal block size (as the smallest after which the plot is (roughly) stable).

Examples

Run this code

# NOT RUN {
set.seed(271)
X <- rPar(5e4, shape = 4)
GEV_shape_plot(X)
abline(h = 1/4, lty = 3) # theoretical xi = 1/shape for Pareto
# }