COMPSTAT 2022: Start Registration
View Submission - COMPSTAT2022
A0488
Title: Total positivity of copulas from a Markov kernel perspective Authors:  Marco Tschimpke - Paris Lodron University Salzburg (Austria) [presenting]
Sebastian Fuchs - University of Salzburg (Austria)
Abstract: An alternative way to examine dependence structures consists in checking whether a copula fulfils certain (positive) dependence properties such as stochastically increasingness (SI) or total positivity of order $2$. We investigate total positivity of order $2$ for a copula's Markov kernel (MK-TP2 for short), presumably the strongest dependence property defined for any copula. We mainly examine the MK-TP2 property within the class of Archimedean and Extreme-Value copulas and characterise the dependence property based on the respective generators.