Multivariate Variance Targeting in the BEKK-GARCH Model

Rasmus Søndergaard Pedersen; Anders Rahbek

doi:10.1111/ectj.12019

Multivariate Variance Targeting in the BEKK-GARCH Model

Rasmus Søndergaard Pedersen, Anders Rahbek

Department of Economics

21 Citations (Scopus)

Abstract

In this paper, we consider asymptotic inference in the multivariate BEKK model based on (co)variance targeting (VT). By definition the VT estimator is a two-step estimator and the theory presented is based on expansions of the modified likelihood function, or estimating function, corresponding to these two steps. Strong consistency is established under weak moment conditions, while sixth-order moment restrictions are imposed to establish asymptotic normality. The simulations included indicate that the multivariately induced higher-order moment constraints are necessary.

Translated title of the contribution	Multivariate Variance Targeting in the BEKK-GARCH Model
Original language	English
Journal	Econometrics Journal
Volume	17
Issue number	1
Pages (from-to)	24-55
ISSN	1368-4221
DOIs	https://doi.org/10.1111/ectj.12019
Publication status	Published - Feb 2014

Keywords

Faculty of Social Sciences
Covariance targeting
Variance targeting
Multivariate GARCH
BEKK
Asymptotic theory
Time series

Access to Document

10.1111/ectj.12019

3 Organisation of and participation in conference
1 Participation in workshop, seminar, course

67th European Meeting of the Econometric Society
Pedersen, R. S. (Speaker)
26 Aug 2013 → 30 Aug 2013
Activity: Participating in or organising an event types › Organisation of and participation in conference
6th Annual SoFiE Conference
Pedersen, R. S. (Speaker)
12 Jun 2013 → 14 Jun 2013
Activity: Participating in or organising an event types › Organisation of and participation in conference
3rd Humboldt-Copenhagen Conference on Financial Econometrics
Pedersen, R. S. (Speaker)
14 Mar 2013 → 16 Mar 2013
Activity: Participating in or organising an event types › Organisation of and participation in conference
PhD Seminar
Pedersen, R. S. (Speaker)
6 Mar 2013
Activity: Participating in or organising an event types › Participation in workshop, seminar, course

Cite this

@article{2b2bd79da9cc4b36a94d5ea0b021423b,

title = "Multivariate Variance Targeting in the BEKK-GARCH Model",

abstract = "In this paper, we consider asymptotic inference in the multivariate BEKK model based on (co)variance targeting (VT). By definition the VT estimator is a two-step estimator and the theory presented is based on expansions of the modified likelihood function, or estimating function, corresponding to these two steps. Strong consistency is established under weak moment conditions, while sixth-order moment restrictions are imposed to establish asymptotic normality. The simulations included indicate that the multivariately induced higher-order moment constraints are necessary.",

keywords = "Faculty of Social Sciences, Covariance targeting, Variance targeting, Multivariate GARCH, BEKK, Asymptotic theory, Time series",

author = "Pedersen, {Rasmus S{\o}ndergaard} and Anders Rahbek",

year = "2014",

month = feb,

doi = "10.1111/ectj.12019",

language = "English",

volume = "17",

pages = "24--55",

journal = "Econometrics Journal",

issn = "1368-4221",

publisher = "Wiley",

number = "1",

}

TY - JOUR

T1 - Multivariate Variance Targeting in the BEKK-GARCH Model

AU - Pedersen, Rasmus Søndergaard

AU - Rahbek, Anders

PY - 2014/2

Y1 - 2014/2

N2 - In this paper, we consider asymptotic inference in the multivariate BEKK model based on (co)variance targeting (VT). By definition the VT estimator is a two-step estimator and the theory presented is based on expansions of the modified likelihood function, or estimating function, corresponding to these two steps. Strong consistency is established under weak moment conditions, while sixth-order moment restrictions are imposed to establish asymptotic normality. The simulations included indicate that the multivariately induced higher-order moment constraints are necessary.

AB - In this paper, we consider asymptotic inference in the multivariate BEKK model based on (co)variance targeting (VT). By definition the VT estimator is a two-step estimator and the theory presented is based on expansions of the modified likelihood function, or estimating function, corresponding to these two steps. Strong consistency is established under weak moment conditions, while sixth-order moment restrictions are imposed to establish asymptotic normality. The simulations included indicate that the multivariately induced higher-order moment constraints are necessary.

KW - Faculty of Social Sciences

KW - Covariance targeting

KW - Variance targeting

KW - Multivariate GARCH

KW - BEKK

KW - Asymptotic theory

KW - Time series

U2 - 10.1111/ectj.12019

DO - 10.1111/ectj.12019

M3 - Journal article

SN - 1368-4221

VL - 17

SP - 24

EP - 55

JO - Econometrics Journal

JF - Econometrics Journal

IS - 1

ER -

Multivariate Variance Targeting in the BEKK-GARCH Model

Abstract

Keywords

Access to Document

Fingerprint

Activities

67th European Meeting of the Econometric Society

6th Annual SoFiE Conference

3rd Humboldt-Copenhagen Conference on Financial Econometrics

PhD Seminar

Cite this