Multithreading for high performance finance risk analysis

Abstract

With the increasing of risks in the financial market, the models of risk management are developing quickly. The standard of the accuracy and effect of the models is improved continuously. This thesis investigates Value at Risk (VaR) which is an important method for measuring the market risk. It reviews the three methods which can be used to quantify VaR. These methods are parameter method, historical data processing method and Monte Carlo simulation method. Monte Carlo simulation has been widely employed for finance risk analysis. One challenge in Monte Carlo simulation is its computation complexity. For this purpose, this thesis researches into multithreading technique for high performance.