Volatility models, risk management and Value-at-Risk models

Value-at-Risk (Var) theory has been the new benchmark for managing financial risk for the last 10 years or so. This project is particularly focused on the empricial study of VaR models. The objective is to study the modeling behaviour of these models in turbulent market times. Data on market indices are to be used to undertake stress tests. Different statistical distributions that have been adapted to the specificites of trading and investment portfolios, will be taken into account. That way, we will be able to check whether models with classical distributions (Normal and Student) are adequate or not for managing portofolio risks that include both long and short positions. The project also deals with different volatility measures that can be used as inputs in VaR models. Measures of both implied volatility and realised intra-day volatility will be under particular scrutiny. The aim is to find the least worst prediction of future volatility. Further extensions to some portfolio risk management modeling issues will be carried out later on.