|We use two nonparametric measures for extreme value dependence to characterise tail dependence or spillover probabilities in the asymptotic dependence and independence case.
We concentrated on the possible contagion case between Brazil and Russian markets and demonstrated how the tail risk measures may be assessed. These new tools have allowed us to document the asymptotic dependence and independence among stock and bond market returns of these countries. Notice, that the omission of asymptotic independence models in studies of contagion could led to mixed results and especially over-estimation could be possibly substantial in connection with measures of contagion. The empirical findings include a confirmation that extreme value dependence was much stronger in the first period of the study.
|In this paper, we use two non-parametric measures for extreme value dependency to characterise tail dependence of returns in five international stock markets and demonstrate how portfolio tail risk may be assessed. These new tools have allowed us to document, for the first time, the widespread asymptotic independence among stock market returns, a phenomenon that has so far been overlooked in the finance literature. The omission of asymptotic independence models has led to over-estimation, possibly substantial, of portfolio risk. Other empirical findings include a confirmation that extreme value dependence is much stronger in bear markets than in bull markets,