distLexBoot: Bootstrapping uncertainty intervals for return periods

Description

Calculates and plots bootstrap uncertainty intervals for distLextremePlot.

Usage

distLexBoot(dlf, nbest = 3, selection = NULL, truncate = 0, n = 100, prop = 0.8, returnall = FALSE, conf.lev = 0.95, RPs = NULL, plot = TRUE, add = FALSE, log = TRUE, progbars = TRUE, ...)

Arguments

dlf

dlf object, as returned by distLextreme, is passed to distLextremePlot.

nbest

Number of best fitted distribution functions in dlf for which bootstrapping is to be done. Overriden by selection. DEFAULT: 3

selection

Character vector with distribution function names to be used. Suggested to keep this low. DEFAULT: NULL

truncate

Truncation of subsamples, see distLquantile. DEFAULT: 0

Number of subsamples to be processed (computing time increases extraordinarily). DEFAULT: 100

prop

Proportion of sample to be used in each run. DEFAULT: 0.8

returnall

Return all simulations, instead of the aggregate confidence level? DEFAULT: FALSE

conf.lev

Confidence level (Proportion of subsamples within 'confidence interval'). Quantiles extracted from this value are passed to quantileMean. DEFAULT: 0.95

RPs

Return Period vector, by default calculated internally based on log. DEFAULT: NULL

plot

Plot results via distLextremePlot? DEFAULT: TRUE

add

Add to existing plot? DEFAULT: FALSE

log

Plot on a logarithmic axis. DEFAULT: TRUE

progbars

Show progress bar for Monte Carlo simulation? DEFAULT: TRUE

...

Further arguments passed to distLextremePlot

Value

A list with (for each selection) a matrix with confidence intervals for RPs, or if returnall=TRUE, all the simulation results

Details

Has not been thoroughly tested yet. Bootstrapping defaults can probably be improved.

Examples

Run this code


data(annMax)
dlf <- distLextreme(annMax, log=TRUE, selection=c("wak","gum","gev","nor"))
dleB <- distLexBoot(dlf, nbest=4, conf.lev=0.5, n=10) # n low for quick example tests