CMStatistics 2020: Start Registration
View Submission - CFE
Title: A COAALA copula for stock-bond return co-movement: Beware of the beast with four tails Authors:  Anne-Florence Allard - University of Bristol (United Kingdom) [presenting]
Hamza Hanbali - Monash University (Australia)
Kristien Smedts - KU Leuven (Belgium)
Abstract: The COAALA copula allows analysing financial market stability by studying comovement between stocks and government bonds using the information on both the global and the four local tail dependence measures (i.e. dependence between severe movements). Such an encompassing view on stock-bond co-movement has never been obtained convincingly before. Our contribution is twofold. First, we develop a novel copula function (COAALA) that is fully flexible. It accommodates the known features of stock-bond dependence. Also, it allows us to study the responses of each asset to a severe move of the other asset, taking into account potential asymmetries in the likelihood of such responses. This copula comes with closed-form expressions of dependence measures without the computational deficiencies of alternative models. Second, for a set of countries with very similar global dependence, the COAALA unravels major differences in local dependencies that hint at different stabilising abilities of bond markets across these countries.