Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
I fitted the Generalized Pareto distribution (GPD) using the POT package in R.
The fitted object provides shape and scale parameters, but not the location parameter. I need it to evaluate the quantile function with qgpd(p, loc = 0, scale = 1, shape = 0, lower.tail = TRUE, lambda = 0).
I am sure this is very basic question (maybe it is mean value, but from which values - all, or only that which are bigger than threshold?) but I am begginer in this area and help is highly appreciated.
The location parameter is the threshold parameter that you provide to the fitgpd function. For example
library(POT)
set.seed(345)
x <- rgpd(2000, 1, 2, 0.25)
fit <- fitgpd(x, threshold = 1)
fit$fitted.values
# scale shape
#2.0501190 0.2471565
See equations (2.3) and (2.4) in the package vignette for details on the model.