Capturing Volatility Persistence: A Dynamically Complete Realized EGARCH-MIDAS Model

Daniel Borup*, Johan S. Jakobsen

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

We introduce extensions of the Realized Exponential GARCH model (REGARCH) that capture the evident high persistence typically observed in measures of financial market volatility in a tractable fashion. The extensions decompose conditional variance into a short-term and a long-term component. The latter utilizes mixed-data sampling or a heterogeneous autoregressive structure, avoiding parameter proliferation otherwise incurred by using the classical ARMA structures embedded in the REGARCH. The proposed models are dynamically complete, facilitating multi-period forecasting. A thorough empirical investigation with an exchange-traded fund that tracks the S&P500 Index and 20 individual stocks shows that our models better capture the dependency structure of volatility. This leads to substantial improvements in empirical fit and predictive ability at both short and long horizons relative to the original REGARCH. A volatility-timing trading strategy shows that capturing volatility persistence yields substantial utility gains for a mean–variance investor at longer investment horizons.
Original languageEnglish
JournalQuantitative Finance
Volume19
Issue number11
Pages (from-to)1839-1855
Number of pages17
ISSN1469-7688
DOIs
Publication statusPublished - Nov 2019

Bibliographical note

Published online: 10. June 2019

Keywords

  • Realized exponential GARCH
  • Persistence
  • Long memory
  • GARCH-MIDAS
  • HAR
  • Realized kernel

Cite this

@article{90a1fbe6a33349ecbcbe0a64a606ae4d,
title = "Capturing Volatility Persistence: A Dynamically Complete Realized EGARCH-MIDAS Model",
abstract = "We introduce extensions of the Realized Exponential GARCH model (REGARCH) that capture the evident high persistence typically observed in measures of financial market volatility in a tractable fashion. The extensions decompose conditional variance into a short-term and a long-term component. The latter utilizes mixed-data sampling or a heterogeneous autoregressive structure, avoiding parameter proliferation otherwise incurred by using the classical ARMA structures embedded in the REGARCH. The proposed models are dynamically complete, facilitating multi-period forecasting. A thorough empirical investigation with an exchange-traded fund that tracks the S&P500 Index and 20 individual stocks shows that our models better capture the dependency structure of volatility. This leads to substantial improvements in empirical fit and predictive ability at both short and long horizons relative to the original REGARCH. A volatility-timing trading strategy shows that capturing volatility persistence yields substantial utility gains for a mean–variance investor at longer investment horizons.",
keywords = "Realized exponential GARCH, Persistence, Long memory, GARCH-MIDAS, HAR, Realized kernel, Realized exponential GARCH, Persistence, Long memory, GARCH-MIDAS, HAR, Realized kernel",
author = "Daniel Borup and Jakobsen, {Johan S.}",
note = "Published online: 10. June 2019",
year = "2019",
month = "11",
doi = "10.1080/14697688.2019.1614653",
language = "English",
volume = "19",
pages = "1839--1855",
journal = "Quantitative Finance",
issn = "1469-7688",
publisher = "Routledge",
number = "11",

}

Capturing Volatility Persistence : A Dynamically Complete Realized EGARCH-MIDAS Model. / Borup, Daniel; Jakobsen, Johan S.

In: Quantitative Finance, Vol. 19, No. 11, 11.2019, p. 1839-1855.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Capturing Volatility Persistence

T2 - A Dynamically Complete Realized EGARCH-MIDAS Model

AU - Borup, Daniel

AU - Jakobsen, Johan S.

N1 - Published online: 10. June 2019

PY - 2019/11

Y1 - 2019/11

N2 - We introduce extensions of the Realized Exponential GARCH model (REGARCH) that capture the evident high persistence typically observed in measures of financial market volatility in a tractable fashion. The extensions decompose conditional variance into a short-term and a long-term component. The latter utilizes mixed-data sampling or a heterogeneous autoregressive structure, avoiding parameter proliferation otherwise incurred by using the classical ARMA structures embedded in the REGARCH. The proposed models are dynamically complete, facilitating multi-period forecasting. A thorough empirical investigation with an exchange-traded fund that tracks the S&P500 Index and 20 individual stocks shows that our models better capture the dependency structure of volatility. This leads to substantial improvements in empirical fit and predictive ability at both short and long horizons relative to the original REGARCH. A volatility-timing trading strategy shows that capturing volatility persistence yields substantial utility gains for a mean–variance investor at longer investment horizons.

AB - We introduce extensions of the Realized Exponential GARCH model (REGARCH) that capture the evident high persistence typically observed in measures of financial market volatility in a tractable fashion. The extensions decompose conditional variance into a short-term and a long-term component. The latter utilizes mixed-data sampling or a heterogeneous autoregressive structure, avoiding parameter proliferation otherwise incurred by using the classical ARMA structures embedded in the REGARCH. The proposed models are dynamically complete, facilitating multi-period forecasting. A thorough empirical investigation with an exchange-traded fund that tracks the S&P500 Index and 20 individual stocks shows that our models better capture the dependency structure of volatility. This leads to substantial improvements in empirical fit and predictive ability at both short and long horizons relative to the original REGARCH. A volatility-timing trading strategy shows that capturing volatility persistence yields substantial utility gains for a mean–variance investor at longer investment horizons.

KW - Realized exponential GARCH

KW - Persistence

KW - Long memory

KW - GARCH-MIDAS

KW - HAR

KW - Realized kernel

KW - Realized exponential GARCH

KW - Persistence

KW - Long memory

KW - GARCH-MIDAS

KW - HAR

KW - Realized kernel

UR - https://sfx-45cbs.hosted.exlibrisgroup.com/45cbs?url_ver=Z39.88-2004&url_ctx_fmt=info:ofi/fmt:kev:mtx:ctx&ctx_enc=info:ofi/enc:UTF-8&ctx_ver=Z39.88-2004&rfr_id=info:sid/sfxit.com:azlist&sfx.ignore_date_threshold=1&rft.object_id=110994087288000&rft.object_portfolio_id=&svc.holdings=yes&svc.fulltext=yes

U2 - 10.1080/14697688.2019.1614653

DO - 10.1080/14697688.2019.1614653

M3 - Journal article

AN - SCOPUS:85067556773

VL - 19

SP - 1839

EP - 1855

JO - Quantitative Finance

JF - Quantitative Finance

SN - 1469-7688

IS - 11

ER -