The Empirical Study Of Monte Carlo Simulation In Risk Measurement

VaR method is an important content of risk management, which has been widely applied in scientific research and empirical analysis. Although the Monte Carlo simulation is one of the key important methods to calculate Va R, the direct application of the traditional Monte Carlo method can result in a conclusion incompatible with reality and even a significant bias. Thus, it is great of theoretical and practical significance to improve the traditional Monte Carlo simulation of calculating Va R. For the questions above, this thesis improved the traditional Monte Carlo simulation of calculating Va R, and the empirical research shows that the prediction effect of the improved method is better than the traditional methods.Specifically, this thesis mainly discuss the following several aspects.Firstly, the background, meaning, domestic and foreign related research of the thesis are introduced, the definition of Va R and its calculation under general distribution are also introduced; What’s more, the historical simulation method,analytic method and Monte Carlo simulation method of calculating Va R are discussed.Secondly, the improved Monte Carlo simulation that combined with SV model is proposed based on the analysis of general Monte Carlo simulation method of calculating Va R.Thirdly, this thesis takes Gem index as the research object and uses improved Monte Carlo simulation method to study Gem index, calculates and analyzes the accuracy of Va R value, and compares the results of Va R with the general Monte Carlo simulation method and the MC-GARCH-Va R method. In order to prevent the impact of the peak and fat tail, the SV-T model is introduced in the Monte Carlo simulation method to calculating Va R value. On the whole, the improved Monte Carlo Simulation method has lots of obvious advantages over other methods with strong feasibility.Finally, we make a brief summary of the thesis, and give some prospects for the problems of further study.