Efficient quantile estimation via a combination of importance sampling and Latin hypercube sampling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many application areas employ a quantile, also known as a percentile or value-at-risk, to measure risk of a stochastic system. We present efficient Monte Carlo methods to estimate a quantile through a combination of importance sampling and Latin hypercube sampling. We also give numerical results from a simple model showing that the combined methods can outperform each by itself.

Original languageEnglish (US)
Title of host publication31st Annual European Simulation and Modelling Conference 2017, ESM 2017
EditorsPaulo J.S. Goncalves
PublisherEUROSIS
Pages49-53
Number of pages5
ISBN (Electronic)9789492859006
StatePublished - 2017
Event31st Annual European Simulation and Modelling Conference, ESM 2017 - Lisbon, Portugal
Duration: Oct 25 2017Oct 27 2017

Publication series

Name31st Annual European Simulation and Modelling Conference 2017, ESM 2017

Other

Other31st Annual European Simulation and Modelling Conference, ESM 2017
Country/TerritoryPortugal
CityLisbon
Period10/25/1710/27/17

All Science Journal Classification (ASJC) codes

  • Modeling and Simulation

Keywords

  • Risk analysis
  • Value-at-risk
  • Variance reduction

Fingerprint

Dive into the research topics of 'Efficient quantile estimation via a combination of importance sampling and Latin hypercube sampling'. Together they form a unique fingerprint.

Cite this