B1113
Title: Cross-dynamical analysis of the tail dependence structures of classic and fuzzy clusters of stocks
Authors: Giovanni De Luca - University of Naples Parthenope (Italy) [presenting]
Paola Zuccolotto - University of Brescia (Italy)
Abstract: A dynamic clustering of financial returns time-series is performed based on the estimated bivariate lower tail dependence coefficients. The resulting groups are composed of assets characterized by high association between extremely low values. The tail dependence coefficients are estimated using the Joe-Clayton copula function. Then, the dynamics of the tail dependence structure of each group (measured e.g. by the average tail dependence) is analyzed in order to identify the group(s) able to anticipate the dynamics of the tail dependence structure of the remaining groups. The analysis is carried out considering both a deterministic and a fuzzy cluster analysis.
Journal