Methods for computing numerical standard errors: Review and application to value-at-risk estimation

David Ardia, Keven Bluteau, Lennart F. Hoogerheide

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Numerical standard error (NSE) is an estimate of the standard deviation of a simulation result if the simulation experiment were to be repeated many times. We review standard methods for computing NSE and perform a Monte Carlo experiments to compare their performance in the case of high/extreme autocorrelation. In particular, we propose an application to risk management where we assess the precision of the value-at-risk measure when the underlying risk model is estimated by simulation-based methods. Overall, heteroscedasticity and autocorrelation estimators with prewhitening perform best in the presence of large/extreme autocorrelation.

LanguageEnglish
Article number20170011
JournalJournal of Time Series Econometrics
Volume10
Issue number2
DOIs
Publication statusPublished - 26 Jul 2018

Fingerprint

Autocorrelation
Value at risk
Standard error
Simulation
Estimator
Monte Carlo experiment
Heteroscedasticity
Risk management
Risk model
Simulation experiment
Standard deviation
Risk measures

Keywords

  • bootstrap
  • GARCH
  • HAC kernel
  • Markov chain Monte Carlo (MCMC)
  • Monte Carlo
  • numerical standard error (NSE)
  • spectral density
  • value-at-risk
  • Welch

Cite this

@article{2e4db7f84658469ba9a293b58796a6a2,
title = "Methods for computing numerical standard errors: Review and application to value-at-risk estimation",
abstract = "Numerical standard error (NSE) is an estimate of the standard deviation of a simulation result if the simulation experiment were to be repeated many times. We review standard methods for computing NSE and perform a Monte Carlo experiments to compare their performance in the case of high/extreme autocorrelation. In particular, we propose an application to risk management where we assess the precision of the value-at-risk measure when the underlying risk model is estimated by simulation-based methods. Overall, heteroscedasticity and autocorrelation estimators with prewhitening perform best in the presence of large/extreme autocorrelation.",
keywords = "bootstrap, GARCH, HAC kernel, Markov chain Monte Carlo (MCMC), Monte Carlo, numerical standard error (NSE), spectral density, value-at-risk, Welch",
author = "David Ardia and Keven Bluteau and Hoogerheide, {Lennart F.}",
year = "2018",
month = "7",
day = "26",
doi = "10.1515/jtse-2017-0011",
language = "English",
volume = "10",
journal = "Journal of Time Series Econometrics",
issn = "2194-6507",
publisher = "de Gruyter",
number = "2",

}

Methods for computing numerical standard errors : Review and application to value-at-risk estimation. / Ardia, David; Bluteau, Keven; Hoogerheide, Lennart F.

In: Journal of Time Series Econometrics, Vol. 10, No. 2, 20170011, 26.07.2018.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Methods for computing numerical standard errors

T2 - Journal of Time Series Econometrics

AU - Ardia, David

AU - Bluteau, Keven

AU - Hoogerheide, Lennart F.

PY - 2018/7/26

Y1 - 2018/7/26

N2 - Numerical standard error (NSE) is an estimate of the standard deviation of a simulation result if the simulation experiment were to be repeated many times. We review standard methods for computing NSE and perform a Monte Carlo experiments to compare their performance in the case of high/extreme autocorrelation. In particular, we propose an application to risk management where we assess the precision of the value-at-risk measure when the underlying risk model is estimated by simulation-based methods. Overall, heteroscedasticity and autocorrelation estimators with prewhitening perform best in the presence of large/extreme autocorrelation.

AB - Numerical standard error (NSE) is an estimate of the standard deviation of a simulation result if the simulation experiment were to be repeated many times. We review standard methods for computing NSE and perform a Monte Carlo experiments to compare their performance in the case of high/extreme autocorrelation. In particular, we propose an application to risk management where we assess the precision of the value-at-risk measure when the underlying risk model is estimated by simulation-based methods. Overall, heteroscedasticity and autocorrelation estimators with prewhitening perform best in the presence of large/extreme autocorrelation.

KW - bootstrap

KW - GARCH

KW - HAC kernel

KW - Markov chain Monte Carlo (MCMC)

KW - Monte Carlo

KW - numerical standard error (NSE)

KW - spectral density

KW - value-at-risk

KW - Welch

UR - http://www.scopus.com/inward/record.url?scp=85050800927&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050800927&partnerID=8YFLogxK

U2 - 10.1515/jtse-2017-0011

DO - 10.1515/jtse-2017-0011

M3 - Article

VL - 10

JO - Journal of Time Series Econometrics

JF - Journal of Time Series Econometrics

SN - 2194-6507

IS - 2

M1 - 20170011

ER -