Multivariate GARCH Models with Impulse Responses in Mean, Variance and Covariance

Ioannidis, Christos; Williams, Julian M.

doi:10.2139/ssrn.957134

Download This Paper

Open PDF in Browser

Add Paper to My Library

Multivariate GARCH Models with Impulse Responses in Mean, Variance and Covariance

16 Pages Posted: 15 Jan 2007

See all articles by Christos Ioannidis

Christos Ioannidis

University of Bath-Department of Economics

Julian M. Williams

Durham Business School

Date Written: 2006

Abstract

This paper presents a unified framework for the computation of impulse response functions for VAR models with MV-GARCH disturbances. By solving from the quadratic form as given by the second moment equations, it is demonstrated that these solutions should be used to adjust the impulse response functions from the mean equations. These adjustments often result in significantly different time profiles for the unadjusted impulse responses. This approach provides for a system consistent solution for multivariate linear autoregressive models, with time varying second moments.

Keywords: MV-GARCH, Impulse Response Functions

JEL Classification: C32, C50

Suggested Citation: Suggested Citation

Ioannidis, Christos and Williams, Julian M., Multivariate GARCH Models with Impulse Responses in Mean, Variance and Covariance (2006). Available at SSRN: https://ssrn.com/abstract=957134 or http://dx.doi.org/10.2139/ssrn.957134