Measuring financial risk : comparison of alternative procedures to estimate VaR and ES

Abstract

We review several procedures for estimating and backtesting two of the most
\nimportant measures of risk, the Value at Risk (VaR) and the Expected
\nShortfall (ES). The alternative estimators differ in the way the specify and
\nestimate the conditional mean and variance and the conditional distribution
\nof returns. The results are illustrated by estimating the VaR and ES of
\ndaily S&P500 returns.