bootfuns: MLE's GPD estimator for xi and boostrap evaluations
In proto4426/PissoortThesis: Thesis in Extreme Value Theory

Description Usage Arguments Details Value Author(s) Examples

Function to compute directly our Maximum Likelihood Estimator (MLE) for the shape parameter of the Generalized Pareto Distribution. It isalso possible to determine bootstrapped confidence interals, either by simple or by double bootstrap

mle.ksiFun(x)

sdmle.ksiFun(x)

ciBootFunc(x, B, level = 0.05, print = T)

ci2Boot.tFunc(x, B1, B2, level = 0.05, print = T)

cov.ci1.vec(n.vec = n.vec1, M, level = 0.05)

`x`	A numeric vector of data to be fitted.
`B`	number of bootstrap replicates for simple bootstrap ci
`level`	confidence level for the confidence intervals
`B1`	number of bootstrap replicates for double bootstrap ci (2) for the first stage
`B1`	number of bootstrap replicates for double bootstrap ci (2) for the second stage

This function is useful to decrease the amount of code and compute directly these estimates.

MLE computed for the GPD or the standard deviation of this MLE

Antoine Pissoort, antoine.pissoort@student.uclouvain.be

X <- above_ured$exceed_red
ksi.chap <- gpd_varu_red$mle[2]
sd.ksi.chap <- sdmle.ksiFun(X)
c500 <- ciBootFunc(B = 500)   # 9 sec.
cc <- ci2Boot.tFunc(B1=100,B2=34)