Abstract
The financial turmoil has aroused the need for risk management tools. Value at Risk (VaR) has been used by many financial institutions to measure market risk and determine the required capital. This thesis analyzes some methods estimating multiple period VaR. In addition to some common approaches like RiskMetrics and empirical distribution, quantile regression is examined. A simulation study is carried out to compare the percentiles calculated by the traditional least square regression model and quantile regression. In real data application, the VaR estimation approaches are applied to the aggregate returns of some major market. The empirical tests indicate that the quantile regression approach is comparable to the other two approaches.