Value-at-risk estimation using a copula approach

MICOCCI, MARCO;MASALA, GIOVANNI BATISTA

2004-01-01

---

Abstract

In this paper we calculate value at risk (VAR) for a two risky assets portfolio assuming that the dependence structure between the two risk factors may be modelled properly by an Archimedean copula function. We then compare our results with the empirical evidence and with more traditional methods of calculation to show that the copula approach gives more reliable results. Key words: copula functions, value at risk, market risk, Monte Carlo simulation, backtesting. JEL CLASSIFICATION: C15, C63, G11, G12, G31, G33 MSC2000 CLASSIFICATION: 91B28, 91B30, 91B70