The univariate models we employed above do not allow for time-varying
conditional covariance between the country and world returns. If the
conditional covariance changes systematically with the introduction of
stock index futures, then our previous results may be biased.
In this section, we will address this problem by estimating the joint
dynamics of each country’s return with the world-market return in a multivariate
GARCH framework that allows for time-varying conditional covariance.
Because we wished to capture the dynamic interaction between
world-market volatility, country-specific volatility, and conditional covariance,
we used the BEKK specification of Engle and Kroner (1995).9
Unlike certain other well-known multivariate GARCH models, the BEKK
model allows conditional variances and covariances to influence each
other.
conditional covariance between the country and world returns. If the
conditional covariance changes systematically with the introduction of
stock index futures, then our previous results may be biased.
In this section, we will address this problem by estimating the joint
dynamics of each country’s return with the world-market return in a multivariate
GARCH framework that allows for time-varying conditional covariance.
Because we wished to capture the dynamic interaction between
world-market volatility, country-specific volatility, and conditional covariance,
we used the BEKK specification of Engle and Kroner (1995).9
Unlike certain other well-known multivariate GARCH models, the BEKK
model allows conditional variances and covariances to influence each
other.