A new estimator for the tail-dependence coefficient

TL;DR

This paper introduces a new, simple estimator for the tail-dependence coefficient in financial risk analysis, demonstrating its consistency, normality, and effectiveness through simulations and real data application.

Contribution

The paper proposes a novel estimator for the tail-dependence coefficient that overcomes previous estimation drawbacks, with proven statistical properties.

Findings

Estimator is strongly consistent and asymptotically normal.

Simulation results show improved finite sample performance.

Application to financial data demonstrates practical utility.

Abstract

Recently, the concept of tail dependence has been discussed in financial applications related to market or credit risk. The multivariate extreme value theory is a proper tool to measure and model dependence, for example, of large loss events. A common measure of tail dependence is given by the so-called tail-dependence coefficient. We present a simple estimator of this latter that avoids the drawbacks of the estimation procedure that has been used so far. We prove strong consistency and asymptotic normality and analyze the finite sample behavior through simulation. We illustrate with an application to financial data.