EMV: Estimation of GPD's parameters by Maximum Likelihood
In genostats/tail.modeling: Estimation of small p-values

Description Usage Arguments Value Author(s) References Examples

Estimates the two parameters k and a of Pareto's cumulative distribution function on a given data set z using the maximum of likelihood.

EMV(z)

`z`	Data set - sorted list of real numbers.

Returns a list of the estimators k and a of the parameters, as real numbers.

Marion

R. M. Hosking and J. R. Wallis, Parameter and Quantile Estimation for the Generalized Pareto Distribution, Technometrics. Available at http://www.jstor.org/stable/1269343.

EMV(tail(sort(rnorm(1e6)),500))

## The function is currently defined as
function (z)
{
  u <- optimize(function(u) Lp(u,z), c(-10, min(1/z)), maximum=TRUE )$maximum
  if(u == 0)
    return(list( a = mean(z), k = 0))
  k <- k_func(u, z)
  list(a = k/u, k = k)
  }