The Research Model Of Risk Measurement On Hushen 300 Index

At present the risk measurement of China's stock index futures is still in the early stage.Due to the lack of data,there are some difficulties in the risk measurement of China's stock index futures.Hushen 300 index is the under-lying index and it has large amount of data. In order to measure the risk of China's stock index futures, we can measure the risk of Hushen 300 index to indirectly estimate the risk. VaR(Value at Risk) is a measuring tool for mar-ket risk,which is widespread used in the world.This thesis empirically studies several topies about measuring the risk of Hushen 300 index with VaR.The research mainly includes three parts:1.Testing the sample data of Hushen 300 index by Basic statistics,Normality Test,Randomness Test and Stationarity Test. We can find that the sample data has high peaks and fat tails properties.2.VaR of Hushen 300 index is estimated by Historical Simulation Method,Normal Model Method,Laplace Model Method, Monte Carlo Simulation Method,Extreme Value Theory. We use different time interval in order to calculate VaR in each kind of method.3.This paper adopts Kupiec statistics as a measure of model effect testing standards.We can find that using Laplace Model Method, Monte Carlo Simu-lation Method and Extreme Value Theory to estimate VaR works best, using Normal Model Method works worst.And using 1-2 years of data modeling is more appropriate.