Essays on modeling international stock market volatility spillovers using multivariate GARCH models

This thesis contributes four essays to the economic literature on the multivariate modeling of the variance-covariance dynamics in the financial time series data, covering the issues of dynamic hedging, international volatility spillovers and financial integration, where applying the Multivariate GARCH models leads to the efficient resolution of a number of ongoing empirical issues. The first essay addresses the portfolio hedging problem for a range of commodities and proposes an alternative way of accounting the asymmetric effect in the volatility by including the asymmetric term in a multivariate GARCH model. By in- and out-of-sample forecast we show that hedging using the asymmetric model outperforms other alternatives. The second essay aims to investigate the cross-border effect of the monetary policy rate announcements on the comovements of the world's major stock markets. The news component of the announcements is extracted from the interest rate futures data. Both the effect of the magnitude and sign of the news are considered. It is demonstrated how these effects may alter the asset allocation decisions. The third essay contributes to the better understanding of the financial integration and volatility, i.e. the risk, transmission across European stock markets and provides comprehensive evidence by using a fully flexible multivariate model. The results are of high practical importance, since the high degree of integration implies higher vulnerability of a country to external risks. On the other, they have considerable implications for financial market participants who are involved in hedging and/or portfolio diversification.(POLS - Sciences politiques et sociales) -- UCL, 201