Fisher Confidence Interval {POT}R Documentation

Fisher Based Confidence Interval for the GP Distribution

Description

Compute Fisher based confidence intervals on parameter and return level for the GP distribution. This is achieved through asymptotic theory and the Observed information matrix of Fisher and eventually the Delta method.

Usage

gpd.fishape(fitted, conf = 0.95)
gpd.fiscale(fitted, conf = 0.95)
gpd.firl(fitted, prob, conf = 0.95)

Arguments

fitted R object given by function fitgpd.
prob The probability of non exceedance.
conf Numeric. The confidence level.

Value

Returns a vector of the lower and upper bound for the confidence interval.

Author(s)

Mathieu Ribatet

See Also

rp2prob, prob2rp, link{gpd.pfscale}, link{gpd.pfshape}, link{gpd.pfrl} and link{confint}

Examples

data(ardieres)
ardieres <- clust(ardieres, 4, 10 / 365, clust.max = TRUE)
fitted <- fitgpd(ardieres[,"obs"], 5, 'mle')
gpd.fishape(fitted)
gpd.fiscale(fitted)

[Package POT version 1.0-9 Index]