Value at risk: Recent advances (Q2702488)

The essence of Value-at-Risk (VaR) modeling is the prediction of the highest expected loss for a given portfolio. The traditional approaches to VaR computations do not provide a satisfactory evaluation of possible losses. The authors review the recent advances in the VaR methodologies and note that the proposed improvements still lack a convincing unified technique capturing the observed phenomena in financial data such as heavy-tails, time-varying volatility and short- and long-range dependence. The authors suggest to use Paretian distributions in VaR modeling.NEWLINENEWLINEFor the entire collection see [Zbl 0954.65001].