CFE 2020: Start Registration
View Submission - CMStatistics
B0402
Title: TailCoR: A new measure of dependence Authors:  Sladana Babic - Ghent University (Belgium) [presenting]
Christophe Ley - University of Luxembourg (Luxembourg)
Lorenzo Ricci - European Stability Mechanism (Luxembourg)
David Veredas - Vlerick Business School (Belgium)
Abstract: Economic and financial crises are characterized by unusually large events. These tail events co-move because of linear and/or nonlinear dependencies. We introduce TailCoR, a metric that combines (and disentangles) these linear and non-linear dependencies. TailCoR between two variables is based on projecting them into a line and computing a tail Inter Quantile Range of the projection. TailCoR is therefore simple to compute. Moreover, no optimizations are needed, it is dimension-free, and it performs well in small samples.