Multivariate Leverage Effects and Realized Semicovariance GARCH Models

Tim Bollerslev; Andrew J. Patton; Rogier Quaedvlieg

doi:10.1016/j.jeconom.2019.12.011

Multivariate Leverage Effects and Realized Semicovariance GARCH Models

Tim Bollerslev, Andrew J. Patton, Rogier Quaedvlieg

Department of Economics and Business Economics - CREATES

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

22 Citations (Scopus)

Abstract

We propose new asymmetric multivariate volatility models. The models exploit estimates of variances and covariances based on the signs of high-frequency returns, measures known as realized semivariances, semicovariances, and semicorrelations, to allow for more nuanced responses to positive and negative return shocks than threshold “leverage effect” terms traditionally used in the literature. Our empirical implementations of the new models, including extensions of widely-used bivariate GARCH specifications for a number of individual stocks and the aggregate market portfolio as well as larger dimensional dynamic conditional correlation type formulations for a cross-section of individual stocks, provide clear evidence of improved model fit and reveal new and interesting asymmetric joint dynamic dependencies.

Original language	English
Journal	Journal of Econometrics
Volume	217
Issue	2
Pages (from-to)	411-430
Number of pages	20
ISSN	0304-4076
DOIs	https://doi.org/10.1016/j.jeconom.2019.12.011
Publication status	Published - Aug 2020

Keywords

Asymmetric dependence
High-frequency data
Realized correlation
Realized volatility
Semivariance

Access to Document

10.1016/j.jeconom.2019.12.011

Cite this

@article{1de280d4989f4e56a929437a75f59439,

title = "Multivariate Leverage Effects and Realized Semicovariance GARCH Models",

abstract = "We propose new asymmetric multivariate volatility models. The models exploit estimates of variances and covariances based on the signs of high-frequency returns, measures known as realized semivariances, semicovariances, and semicorrelations, to allow for more nuanced responses to positive and negative return shocks than threshold “leverage effect” terms traditionally used in the literature. Our empirical implementations of the new models, including extensions of widely-used bivariate GARCH specifications for a number of individual stocks and the aggregate market portfolio as well as larger dimensional dynamic conditional correlation type formulations for a cross-section of individual stocks, provide clear evidence of improved model fit and reveal new and interesting asymmetric joint dynamic dependencies.",

keywords = "Asymmetric dependence, High-frequency data, Realized correlation, Realized volatility, Semivariance",

author = "Tim Bollerslev and Patton, {Andrew J.} and Rogier Quaedvlieg",

year = "2020",

month = aug,

doi = "10.1016/j.jeconom.2019.12.011",

language = "English",

volume = "217",

pages = "411--430",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Multivariate Leverage Effects and Realized Semicovariance GARCH Models

AU - Bollerslev, Tim

AU - Patton, Andrew J.

AU - Quaedvlieg, Rogier

PY - 2020/8

Y1 - 2020/8

N2 - We propose new asymmetric multivariate volatility models. The models exploit estimates of variances and covariances based on the signs of high-frequency returns, measures known as realized semivariances, semicovariances, and semicorrelations, to allow for more nuanced responses to positive and negative return shocks than threshold “leverage effect” terms traditionally used in the literature. Our empirical implementations of the new models, including extensions of widely-used bivariate GARCH specifications for a number of individual stocks and the aggregate market portfolio as well as larger dimensional dynamic conditional correlation type formulations for a cross-section of individual stocks, provide clear evidence of improved model fit and reveal new and interesting asymmetric joint dynamic dependencies.

AB - We propose new asymmetric multivariate volatility models. The models exploit estimates of variances and covariances based on the signs of high-frequency returns, measures known as realized semivariances, semicovariances, and semicorrelations, to allow for more nuanced responses to positive and negative return shocks than threshold “leverage effect” terms traditionally used in the literature. Our empirical implementations of the new models, including extensions of widely-used bivariate GARCH specifications for a number of individual stocks and the aggregate market portfolio as well as larger dimensional dynamic conditional correlation type formulations for a cross-section of individual stocks, provide clear evidence of improved model fit and reveal new and interesting asymmetric joint dynamic dependencies.

KW - Asymmetric dependence

KW - High-frequency data

KW - Realized correlation

KW - Realized volatility

KW - Semivariance

UR - http://www.scopus.com/inward/record.url?scp=85078081499&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2019.12.011

DO - 10.1016/j.jeconom.2019.12.011

M3 - Journal article

SN - 0304-4076

VL - 217

SP - 411

EP - 430

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 2

ER -

Multivariate Leverage Effects and Realized Semicovariance GARCH Models

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this