A conditional extreme value volatility estimator based on high-frequency returns

Turan G. Bali, David Weinbaum

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

This paper introduces a conditional extreme value volatility estimator (EVT) based on high-frequency returns. The relative performance of the EVT is compared with the discrete-time GARCH and implied volatility models for 1-day and 20-day-ahead forecasts of realized volatility. This is also a first attempt towards detecting any time-series variation in extreme value distributions using high-frequency intraday data. The information content of EVT is examined in the context of forecasting S&P 100 index volatility. Adjusted-R2 values imply superior performance of the implied volatility index (VIX) and EVT in capturing time-series variation in realized volatility. The forecasting ability of various discrete-time GARCH models turns out to be inferior to VIX and EVT. According to the Theil inequality coefficient and the heteroscedasticity-adjusted root mean squared and mean absolute errors, (1) EVT provides more accurate forecasts than the VIX and GARCH volatility models; (2) VIX generally yields a less accurate characterization of realized volatility than EVT and GARCH models. These results have implications for financial risk management, and are thus relevant to both regulators and practitioners.

Original languageEnglish (US)
Pages (from-to)361-397
Number of pages37
JournalJournal of Economic Dynamics and Control
Volume31
Issue number2
DOIs
StatePublished - Feb 2007
Externally publishedYes

Fingerprint

Extreme Values
Volatility
Realized Volatility
Estimator
Implied Volatility
Time series
GARCH Model
Generalized Autoregressive Conditional Heteroscedasticity
Forecast
Forecasting
Risk management
Coefficient Inequalities
High-frequency Data
Financial Risk
Extreme Value Distribution
Heteroscedasticity
Discrete-time Model
Information Content
Risk Management
Regulator

Keywords

  • Extreme value
  • GARCH
  • High-frequency returns
  • Implied volatility
  • Realized volatility

ASJC Scopus subject areas

  • Economics and Econometrics
  • Control and Optimization

Cite this

A conditional extreme value volatility estimator based on high-frequency returns. / Bali, Turan G.; Weinbaum, David.

In: Journal of Economic Dynamics and Control, Vol. 31, No. 2, 02.2007, p. 361-397.

Research output: Contribution to journalArticle

@article{84d232634ef64fcab75f7288ddf01068,
title = "A conditional extreme value volatility estimator based on high-frequency returns",
abstract = "This paper introduces a conditional extreme value volatility estimator (EVT) based on high-frequency returns. The relative performance of the EVT is compared with the discrete-time GARCH and implied volatility models for 1-day and 20-day-ahead forecasts of realized volatility. This is also a first attempt towards detecting any time-series variation in extreme value distributions using high-frequency intraday data. The information content of EVT is examined in the context of forecasting S&P 100 index volatility. Adjusted-R2 values imply superior performance of the implied volatility index (VIX) and EVT in capturing time-series variation in realized volatility. The forecasting ability of various discrete-time GARCH models turns out to be inferior to VIX and EVT. According to the Theil inequality coefficient and the heteroscedasticity-adjusted root mean squared and mean absolute errors, (1) EVT provides more accurate forecasts than the VIX and GARCH volatility models; (2) VIX generally yields a less accurate characterization of realized volatility than EVT and GARCH models. These results have implications for financial risk management, and are thus relevant to both regulators and practitioners.",
keywords = "Extreme value, GARCH, High-frequency returns, Implied volatility, Realized volatility",
author = "Bali, {Turan G.} and David Weinbaum",
year = "2007",
month = "2",
doi = "10.1016/j.jedc.2005.10.002",
language = "English (US)",
volume = "31",
pages = "361--397",
journal = "Journal of Economic Dynamics and Control",
issn = "0165-1889",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - A conditional extreme value volatility estimator based on high-frequency returns

AU - Bali, Turan G.

AU - Weinbaum, David

PY - 2007/2

Y1 - 2007/2

N2 - This paper introduces a conditional extreme value volatility estimator (EVT) based on high-frequency returns. The relative performance of the EVT is compared with the discrete-time GARCH and implied volatility models for 1-day and 20-day-ahead forecasts of realized volatility. This is also a first attempt towards detecting any time-series variation in extreme value distributions using high-frequency intraday data. The information content of EVT is examined in the context of forecasting S&P 100 index volatility. Adjusted-R2 values imply superior performance of the implied volatility index (VIX) and EVT in capturing time-series variation in realized volatility. The forecasting ability of various discrete-time GARCH models turns out to be inferior to VIX and EVT. According to the Theil inequality coefficient and the heteroscedasticity-adjusted root mean squared and mean absolute errors, (1) EVT provides more accurate forecasts than the VIX and GARCH volatility models; (2) VIX generally yields a less accurate characterization of realized volatility than EVT and GARCH models. These results have implications for financial risk management, and are thus relevant to both regulators and practitioners.

AB - This paper introduces a conditional extreme value volatility estimator (EVT) based on high-frequency returns. The relative performance of the EVT is compared with the discrete-time GARCH and implied volatility models for 1-day and 20-day-ahead forecasts of realized volatility. This is also a first attempt towards detecting any time-series variation in extreme value distributions using high-frequency intraday data. The information content of EVT is examined in the context of forecasting S&P 100 index volatility. Adjusted-R2 values imply superior performance of the implied volatility index (VIX) and EVT in capturing time-series variation in realized volatility. The forecasting ability of various discrete-time GARCH models turns out to be inferior to VIX and EVT. According to the Theil inequality coefficient and the heteroscedasticity-adjusted root mean squared and mean absolute errors, (1) EVT provides more accurate forecasts than the VIX and GARCH volatility models; (2) VIX generally yields a less accurate characterization of realized volatility than EVT and GARCH models. These results have implications for financial risk management, and are thus relevant to both regulators and practitioners.

KW - Extreme value

KW - GARCH

KW - High-frequency returns

KW - Implied volatility

KW - Realized volatility

UR - http://www.scopus.com/inward/record.url?scp=33846318162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846318162&partnerID=8YFLogxK

U2 - 10.1016/j.jedc.2005.10.002

DO - 10.1016/j.jedc.2005.10.002

M3 - Article

AN - SCOPUS:33846318162

VL - 31

SP - 361

EP - 397

JO - Journal of Economic Dynamics and Control

JF - Journal of Economic Dynamics and Control

SN - 0165-1889

IS - 2

ER -