atSite: At-site analysis

In martindurocher/floodRFA: A R-package for regional frequency of floods

Description

Return the fitting of the best distribution according to AIC/BIC among an initial set. This function is a wrapping of the functions found in the package lmomco-package. The function atSite2par extract the parameters in the format of function vec2par. The function update refit the distribution using the new parameters passed in arguments.

Usage

atSite(x, distr = c("gev", "glo", "pe3", "ln3"), method = "lmom", k = 2,
  warn = FALSE, tol.gev = 0, rm.zero = FALSE)

atSite2par(obj)

## S3 method for class 'atSite'
update(obj, data = NULL, method = NULL, type = NULL,
  k = NULL, tol.gev = NULL)

Arguments

x	Vector of a sample.
distr	List of strings that represents candidates distributions to fit. By default: 'gev', 'glo', 'pe3', 'gno'. See vec2par for the list of available distribution.
method	String that determines the estimation method. Either maximum likelihood ('mle') or L-moments ('lmom').
k	Penalty factor for criterion k*p - 2*L, where p is the number of parameter and L is the log-likelihood. If k = 2 (default) the criterion is the AIC and if k = log(n) the criterion is the BIC.
tol.gev	Numerical threshold representing the accepted difference between criteria to consider similar. In that case the GEV distribution is preferred over the best distribution identified. @param rm.zero Logical. Should the zero value be removed.

References

Coles, S. (2001). An introduction to statistical modeling of extreme values. Springer Verlag.

Hosking, J. R. M., & Wallis, J. R. (1997). Regional frequency analysis: an approach based on L-moments. Cambridge Univ Pr.

See Also

mle2par, lmom2par.

Examples

library(lmomco)
x <- rlmomco(49, vec2par(c(100,20,.1),'gev'))

ax <- atSite(x)

print(ax)

ax2 <- update(ax, method = 'mle', tol.gev = 2)

martindurocher/floodRFA documentation built on June 5, 2019, 8:44 p.m.