O'Reilly logo

Analysis of Financial Time Series, Third Edition by RUEY S. TSAY

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

10.4 GARCH Models for Bivariate Returns

Since the same techniques can be used to generalize many univariate volatility models to the multivariate case, we focus our discussion on the multivariate GARCH model. Other multivariate volatility models can also be used.

For a k-dimensional return series inline, a multivariate GARCH model uses “exact equations” to describe the evolution of the k(k + 1)/2-dimensional vector inline over time. By exact equation, we mean that the equation does not contain any stochastic shock. However, the exact equation may become complicated even in the simplest case of k = 2 for which inline is three dimensional. To keep the model simple, some restrictions are often imposed on the equations.

10.4.1 Constant-Correlation Models

To keep the number of volatility equations low, Bollerslev (1990) considers the special case in which the correlation coefficient ρ21, t = ρ21 is time invariant, where |ρ21| < 1. Under such an assumption, ρ21 is a constant parameter and the volatility model consists of two equations for inline, which is defined as . A GARCH(1,1) model for becomes

10.21

where

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required