Abstract:
Value at Risk (VaR) forecasts have been increasingly accepted globally by both risk managers and regulators as a tool to identify and control exposure to financial market risk. However, modern portfolios are characterized by a constantly changing composition of security holdings that reflect portfolio managers' strategies, expected prices, and net cash flows into the portfolio. As a result of these factors, portfolio returns are time-varying mixtures of distributions which are unlikely to be well approximated by conventional methods.