poolGrp: Perform regional frequency analysis using L-moments on...
In martindurocher/floodRFA: A R-package for regional frequency of floods

Perform regional frequency analysis using L-moments on pooling groups

1 2	poolGrp(obj, dis, rtn = c(2, 5, 10, 20, 50, 100), nk = 25, sset = NULL, indx = NULL, distr = NULL, rm.target = FALSE, target = NULL)

`obj`	Output from rfa. Containt the index-Flood and the Lmoments of all sites
`dis`	If a vector of distances, the target must be included with distance zero. If a distance matrix or a distance object, the target must be specified.
`rtn`	A vector of return period
`nk`	The number of donor
`indx`	A vector or string to force new index-flood value
`distr`	Regioal distribution to use. If omited, a best distribution is selected
`rm.target`	Should the target be removed from the
`nsim`	Number of simulation used to compute Homegeneity and GOF statistics
`nk.min`	Minimal size of pooling groups

library(lmomco)
coord <- replicate(2, runif(100))
dis <- dist(coord)

para <- vec2par(c(100, 30, .1),'gev')
xdata <- lapply(1:100, function(ii) rlmomco(rpois(1,20)+20, para))
names(xdata) <- paste('site',1:100)

obj <- rfa(xdata, 'median')

rfaRegFit(obj)

# Perform the analysis of the site 13th site
regGev <- poolGrp(obj, dis = dis, target = 13, distr = 'gev')
regGev

# Could choose the distribution automatically
poolGrp(obj, dis = dis, target = 13)

# To get information on the regional growth curve
poolGrp(obj, dis = dis, target = 13, indx = 1)