B1693
Title: Copula-based measures for dependence between random vectors
Authors: Steven De Keyser - KU Leuven (Belgium)
Irene Gijbels - KU Leuven (Belgium) [presenting]
Abstract: Copula-based dependence quantification is discussed between multiple groups of random variables, of possibly different sizes. Some recent approaches are briefly commented on, and subsequently, one approach is elaborated. For statistical inference, the focus is on the absolutely continuous setting assuming copula densities exist. Parametric and semi-parametric frameworks are considered, estimation procedures are discussed, and asymptotic properties of the proposed estimators are established. Simulations indicate finite-sample performance and practical use is discussed.
