lmom.hist.fit: L-moment estimation in the presence of Historical Data

In ilapros/ilaprosUtils: A Miscellaneous Collection of Functions I Tend to Use

L-moment estimation in the presence of Historical Data

Description

L-moment fitting procedures for sample of both systematic and historical data via Partial Probability Weighted Moments (Wang, 1990).

Usage

lmom.hist.fit(xdat, k = NULL, h = NULL, X0 = NULL, nmom = 3)

Arguments

xdat	vector of historical and systematic/observed data - the first k elements of the vector should be the historical events
k	number of historical events available. These events should be stored as the first observations of the xdat vector
h	length of years covered by the information of the historical period
X0	the perception threshold which is exceeded by the historical events. This is most likely different from the lowest historical value. It should indicate a value after which we are confident the event would have been recorded or left traces
nmom	the number of moments to estimate

Value

A vector of L-moment, as lmom::samlmu.

References

Wang, Q.J.(1990). Unbiased estimation of probability weighted moments and partial probability weighted moments from systematic and historical flood information and their application. Journal of hydrology, 120, 115–124.

Examples

# library(ismev)
set.seed(54165784)
xx <- rgev(500, 40, 6, -0.2)
xxsist <- xx[471:500]; xxhist <- xx[1:470][xx[1:470] > 80]
lmh <- lmom.hist.fit(c(xxhist,xxsist), k = length(xxhist), h = 470, X0 = 80, nmom = 5)
lmh
lmom::pelgev(lmh)
gev.hist.fit(c(xxhist,xxsist), k = length(xxhist), h = 470, X0 = 80, show = FALSE)$mle
### note the different parametrization for the shape parameter

ilapros/ilaprosUtils documentation built on April 6, 2023, 4:44 a.m.