Fractional integration and fat tails for realized covariance kernels

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

We introduce a new fractionally integrated model for covariance matrix dynamics based on the long-memory behavior of daily realized covariance matrix kernels. We account for fat tails in the data by an appropriate distributional assumption. The covariance matrix dynamics are formulated as a numerically efficient matrix recursion that ensures positive definiteness under simple parameter constraints. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model statistically and economically outperforms recent alternatives such as the multivariate HEAVY model and the multivariate HAR model. In addition, the long-memory behavior is more important during non-crisis periods.

Original languageEnglish
Article number17
Pages (from-to)66-90
Number of pages25
JournalJournal of Financial Econometrics
Volume17
Issue number1
DOIs
Publication statusPublished - 1 Jan 2019

Fingerprint

Kernel
Fractional integration
Fat tails
Covariance matrix
Integrated model
Long memory
Multivariate models
Recursion

Keywords

  • fractional integration
  • heavy tails
  • matrix-F distribution
  • multivariate volatility
  • realized covariance matrices
  • score dynamics

Cite this

@article{45ffed3945df4c13a9d8508c99ae98da,
title = "Fractional integration and fat tails for realized covariance kernels",
abstract = "We introduce a new fractionally integrated model for covariance matrix dynamics based on the long-memory behavior of daily realized covariance matrix kernels. We account for fat tails in the data by an appropriate distributional assumption. The covariance matrix dynamics are formulated as a numerically efficient matrix recursion that ensures positive definiteness under simple parameter constraints. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model statistically and economically outperforms recent alternatives such as the multivariate HEAVY model and the multivariate HAR model. In addition, the long-memory behavior is more important during non-crisis periods.",
keywords = "fractional integration, heavy tails, matrix-F distribution, multivariate volatility, realized covariance matrices, score dynamics",
author = "Anne Opschoor and Andre Lucas",
year = "2019",
month = "1",
day = "1",
doi = "10.1093/jjfinec/nby029",
language = "English",
volume = "17",
pages = "66--90",
journal = "Journal of Financial Econometrics",
issn = "1479-8409",
publisher = "Oxford University Press",
number = "1",

}

Fractional integration and fat tails for realized covariance kernels. / Opschoor, Anne; Lucas, Andre.

In: Journal of Financial Econometrics, Vol. 17, No. 1, 17, 01.01.2019, p. 66-90.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Fractional integration and fat tails for realized covariance kernels

AU - Opschoor, Anne

AU - Lucas, Andre

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We introduce a new fractionally integrated model for covariance matrix dynamics based on the long-memory behavior of daily realized covariance matrix kernels. We account for fat tails in the data by an appropriate distributional assumption. The covariance matrix dynamics are formulated as a numerically efficient matrix recursion that ensures positive definiteness under simple parameter constraints. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model statistically and economically outperforms recent alternatives such as the multivariate HEAVY model and the multivariate HAR model. In addition, the long-memory behavior is more important during non-crisis periods.

AB - We introduce a new fractionally integrated model for covariance matrix dynamics based on the long-memory behavior of daily realized covariance matrix kernels. We account for fat tails in the data by an appropriate distributional assumption. The covariance matrix dynamics are formulated as a numerically efficient matrix recursion that ensures positive definiteness under simple parameter constraints. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model statistically and economically outperforms recent alternatives such as the multivariate HEAVY model and the multivariate HAR model. In addition, the long-memory behavior is more important during non-crisis periods.

KW - fractional integration

KW - heavy tails

KW - matrix-F distribution

KW - multivariate volatility

KW - realized covariance matrices

KW - score dynamics

UR - http://www.scopus.com/inward/record.url?scp=85062358448&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062358448&partnerID=8YFLogxK

U2 - 10.1093/jjfinec/nby029

DO - 10.1093/jjfinec/nby029

M3 - Article

VL - 17

SP - 66

EP - 90

JO - Journal of Financial Econometrics

JF - Journal of Financial Econometrics

SN - 1479-8409

IS - 1

M1 - 17

ER -