UKFE (version 1.0.2)

GEVPars: Generalised extreme value distribution parameter estimates

Description

Estimated parameters from a sample (with Lmoments or maximum likelihood estimation) or from L1 (first L-moment), Lcv (linear coefficient of variation), and LSkew (linear skewness)

Usage

GEVPars(x = NULL, mle = FALSE, L1 = NULL, LCV = NULL, LSKEW = NULL)

Value

Parameter estimates (location, scale, shape)

Arguments

x

numeric vector. The sample

mle

logical argument with a default of FALSE. If FALSE the parameters are estimated with Lmoments, if TRUE the parameters are estimated by maximum likelihood estimation

L1

first Lmoment

LCV

linear coefficient of variation

LSKEW

linear skewness

Author

Anthony Hammond

Details

The L-moment estimated parameters are by the method detailed in 'Hosking J. and Wallis J. 1997 Regional Frequency Analysis: An Approach Based on L-moments. Cambridge University Press, New York'.

This function applies a probability distribution model which assumes that the sample data is independent and identical, i.e. the assumption is that all observations in the sample would not impact or depend on any other. Furthermore, all observations are from the same underlying process which has not changed over the period of record (stationarity).

Examples

# Get an annual maximum sample and estimate the parameters using L-moments
am_27090 <- GetAM(27090)
GEVPars(am_27090$Flow)

# Estimate parameters using MLE
GEVPars(am_27090$Flow, mle = TRUE)

# Calculate L-moments and estimate the parameters with L1, Lcv, and Lskew
LMoments(am_27090$Flow)

# Store L-moments in an object
l_pars <- as.numeric(LMoments(am_27090$Flow))[c(1, 5, 6)]
GEVPars(L1 = l_pars[1], LCV = l_pars[2], LSKEW = l_pars[3])