EPDfit: Fit the Extended Pareto distribution to a vector of...

In freakonometrics/TopIncomes: Estimation of Top Incomes Using Pareto Models

Description

Fit the Extended Pareto distribution to a vector of observations, with weights, using maximum likelihood estimation

Usage

EPDfit(data, tau, start = c(0.1, 1), warnings = FALSE,
  weights = rep(1, length(data)))

Arguments

data	numeric, vector of observations
tau	numeric, value for tau in the EPD distribution
start	numeric, vector of length 2 containing the starting values for the optimisation (default are c(.1,1)
warnings	logical indicating if possible warnings from the optimisation function are shown
weights	numeric, vector of (positive) weights
rho	numeric, parameter of Fraga Alves et al. (2003) estimate

Value

a list with gamma the vector of the corresponding estimates for the tail parameter of the EPD, kappa the vector of the corresponding MLE estimates for the kappa parameter of the EPD, as well as tau

Source

adapted from https://github.com/TReynkens/ReIns/blob/master/R/EPD.R Tom Reynkens, ReIns package version 1.0.7

References

Charpentier & Flachaire (2019) Pareto Models for Top Incomes hal-02145024

Examples

set.seed(123)
x <- rpareto1(100, mu=1, alpha=.5)
w <- rgamma(100,10,10)
EPDfit(data=x, tau=-3.3, weights=w)

freakonometrics/TopIncomes documentation built on July 7, 2019, 8:06 a.m.