Calculating Maximum Likelihood Estimators for the Generalized Pareto Distribution,

Abstract

The Generalized Pareto Distribution (GPD) is a two-parameter family of distributions which can be used to model exceedences over a threshold. Maximum likelihood parameter estimates are preferred since they are asymptotically normal and asymptotically efficient. Numerical methods are required for maximizing the loglikelihood since the minimal sufficient statistics are the order statistics and there is no obvious simplification of the nonlinear likelihood equation. An algorithm is given to compute GPD maximum likelihood estimates by reducing the two-dimensional numerical search for the zeros of the gradient vector to a one-dimensional numerical search.

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1992
Accession Number
ADP007224

Entities

People

  • Scott D. Grimshaw

Organizations

  • University of Maryland

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Computer Science
  • Computing-Related Activities
  • Data Science
  • Engineering
  • Equations
  • Estimators
  • Information Science
  • Interdisciplinary Science
  • Mathematics
  • Network Science
  • Order Statistics
  • Statistics
  • Theoretical Computer Science
  • Two Dimensional

Fields of Study

  • Mathematics

Readers

  • Statistical inference.