Value at Risk Estimation in a Mixture Normality Framework

Value at Risk (VaR) has emerged as a useful tool to risk management. A relevant driving force has been the diffusion as a benchmark of JP Morgan RiskMetricsTM methodology and the subsequent BIS adoption of VaR for all trading portfolios of financial institutions. In this paper we analyze the use of mixture of truncated normal distributions in VaR modelling along with an optimization algorithm to identify the optimal thresholds. The approach gives evidence to capture the extreme tails much better than the standard VaR RiskMetricsTM method completely maintaining local normality properties in the model. Simulation results applied to international equity portfolios are presente