Realized wishart-garch: A score-driven multi-Asset volatility model

P. Gorgi, P. R. Hansen, P. Janus, S. J. Koopman

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

We propose a novel multivariate GARCH model that incorporates realized measures for the covariance matrix of returns. The joint formulation of a multivariate dynamic model for outer-products of returns, realized variances, and realized covariances leads to a feasible approach for analysis and forecasting. The updating of the covariance matrix relies on the score function of the joint likelihood function based on Gaussian and Wishart densities. The dynamic model is parsimonious while the analysis relies on straightforward computations. In a Monte Carlo study, we show that parameters are estimated accurately for different small sample sizes. We illustrate the model with an empirical in-sample and out-of-sample analysis for a portfolio of 15 U.S. financial assets.

Original languageEnglish
Pages (from-to)1-32
Number of pages32
JournalJournal of Financial Econometrics
Volume17
Issue number1
DOIs
Publication statusPublished - Jan 2019

Fingerprint

Assets
Volatility models
Covariance matrix
Small sample
Sample size
Multivariate GARCH models
Realized variance
Monte Carlo study
Financial assets

Keywords

  • high-frequency data
  • multivariate GARCH
  • multivariate volatility
  • realized covariance
  • score
  • Wishart distribution

Cite this

Gorgi, P. ; Hansen, P. R. ; Janus, P. ; Koopman, S. J. / Realized wishart-garch : A score-driven multi-Asset volatility model. In: Journal of Financial Econometrics. 2019 ; Vol. 17, No. 1. pp. 1-32.
@article{ecd69bdb887a4c718369a43ae9d29222,
title = "Realized wishart-garch: A score-driven multi-Asset volatility model",
abstract = "We propose a novel multivariate GARCH model that incorporates realized measures for the covariance matrix of returns. The joint formulation of a multivariate dynamic model for outer-products of returns, realized variances, and realized covariances leads to a feasible approach for analysis and forecasting. The updating of the covariance matrix relies on the score function of the joint likelihood function based on Gaussian and Wishart densities. The dynamic model is parsimonious while the analysis relies on straightforward computations. In a Monte Carlo study, we show that parameters are estimated accurately for different small sample sizes. We illustrate the model with an empirical in-sample and out-of-sample analysis for a portfolio of 15 U.S. financial assets.",
keywords = "high-frequency data, multivariate GARCH, multivariate volatility, realized covariance, score, Wishart distribution",
author = "P. Gorgi and Hansen, {P. R.} and P. Janus and Koopman, {S. J.}",
year = "2019",
month = "1",
doi = "10.1093/jjfinec/nby007",
language = "English",
volume = "17",
pages = "1--32",
journal = "Journal of Financial Econometrics",
issn = "1479-8409",
publisher = "Oxford University Press",
number = "1",

}

Realized wishart-garch : A score-driven multi-Asset volatility model. / Gorgi, P.; Hansen, P. R.; Janus, P.; Koopman, S. J.

In: Journal of Financial Econometrics, Vol. 17, No. 1, 01.2019, p. 1-32.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Realized wishart-garch

T2 - A score-driven multi-Asset volatility model

AU - Gorgi, P.

AU - Hansen, P. R.

AU - Janus, P.

AU - Koopman, S. J.

PY - 2019/1

Y1 - 2019/1

N2 - We propose a novel multivariate GARCH model that incorporates realized measures for the covariance matrix of returns. The joint formulation of a multivariate dynamic model for outer-products of returns, realized variances, and realized covariances leads to a feasible approach for analysis and forecasting. The updating of the covariance matrix relies on the score function of the joint likelihood function based on Gaussian and Wishart densities. The dynamic model is parsimonious while the analysis relies on straightforward computations. In a Monte Carlo study, we show that parameters are estimated accurately for different small sample sizes. We illustrate the model with an empirical in-sample and out-of-sample analysis for a portfolio of 15 U.S. financial assets.

AB - We propose a novel multivariate GARCH model that incorporates realized measures for the covariance matrix of returns. The joint formulation of a multivariate dynamic model for outer-products of returns, realized variances, and realized covariances leads to a feasible approach for analysis and forecasting. The updating of the covariance matrix relies on the score function of the joint likelihood function based on Gaussian and Wishart densities. The dynamic model is parsimonious while the analysis relies on straightforward computations. In a Monte Carlo study, we show that parameters are estimated accurately for different small sample sizes. We illustrate the model with an empirical in-sample and out-of-sample analysis for a portfolio of 15 U.S. financial assets.

KW - high-frequency data

KW - multivariate GARCH

KW - multivariate volatility

KW - realized covariance

KW - score

KW - Wishart distribution

UR - http://www.scopus.com/inward/record.url?scp=85054137869&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054137869&partnerID=8YFLogxK

U2 - 10.1093/jjfinec/nby007

DO - 10.1093/jjfinec/nby007

M3 - Article

VL - 17

SP - 1

EP - 32

JO - Journal of Financial Econometrics

JF - Journal of Financial Econometrics

SN - 1479-8409

IS - 1

ER -