Vine copula : the new approach for modeling high dimensional dependencies : application to financial data

Abstract EN

A popular approach to build a High-dimensional dependence is based on copulas and vine copulas.

The aim of this work is: the construction of several structures of dependence of a set of stock market indices via vine copulas for risk management.

For this, we took as data four stock market indices, namely the cac40, dawJones, Nikkei225, FTSE100.

Each series was modeled separately using the ARMA-GARCH models then we switched to multiple modeling via vine copula.

The structure of the vine was identified using Kendall's τCopulas; vine copula; dependency structure; stock market indices.