Modeling the Conditional Covariance between Stock and Bond Returns : A Multivariate GARCH Approach

To analyze the intertemporal interaction between the stock and bond market returns, we allow the conditional covariance matrix to vary over time according to a multivariate GARCH model similar to Bollerslev, Engle and Wooldridge (1988). We extend the model such that it allows for asymmetric effects on conditional variances and covariances. Using weekly U.S. stock and bond market data, we find strong evidence of conditional heteroskedasticity in the covariance between stock and bond market returns. The results indicate that not only variances, but also covariances respond asymmetrically to return shocks. Regardless of the bond market shocks, bad news in the stock market is typically followed by a higher conditional covariance than good news. We find that volatility timing strategies for dynamic asset allocation significantly outperform passive strategies. Even when short-sale restrictions are present and transaction costs are high, the economic value of dynamic trading strategies is larger than that of a passive strategy. Moreover, the symmetric volatility timing strategy is outperformed by its asymmetric counterpart