China's Open-ended Equity Fund Var Market Risk Measure

VaR(Value at Risk) has been attracting the attentions from the financial institutions and the financial supervision departments around the world since its first show in 1990s. It also has become the main method in the measurement of the financial market risk, and stepped its way in the field of other financial risk management such as credit risk and liquidity risk managements.The open-ended funds started to develop just in recent years in China. 'Hua'an Chuang Xin', the first open-ended fund in China, was established in Sept., 2001, and from then on China's security investment fund started to boom. At the same time, the market risk appears to be increasing gradually, especially for the open-ended equity fund, because the fund units are redeemable and its main investment field is the stock market, where is full of high volatility. So the market risk management for such fund seems more important.This paper summarizes the risk measurement methods in the open-ended equity fund market first, and then gives a detailed introduction to the VaR theory and the GARCH(Autoregressive Conditional Heteroskedasticity) related models. The VaR theory part includes the concept of VaR, the calculation principle and methods(including parameter method, historical simulation and Monte Carlo simulation) and the VaR test method—Kupiec's performance test. GARCH, GARCH-M, TARCH, EGARCH and PARCH models are introduced in this paper to gain the volatility clustering character in the fund return series. The goal is to calculate VaR based on the GARCH related models.In the empirical analysis part, statistical descriptions on the sample funds are made firstly, which show that the actual distribution of each sample fund's daily net value return possesses the characteristic of leptokurtosis. So it is necessary to add student T distribution and GED(Generalized Error Distribution) to capture such leptokurtosis characteristic other than normal distribution. Secondly, ARCH test shows that there exists volatility clustering in each sample fund's daily net value return, so GARCH related models should be used to describe such volatility clustering characteristic. After model selection, the GARCH(1,1) model is the best model to describe the daily net value return's volatility of the open-ended equity funds in China.Then, under the hypothesis of normal distribution, student T distribution and GED, based on the GARCH(1,1) model, choosing confidence level of 95% and 99%, VaR is calculated by the parameter method and Monte Carlo simulation method, and Kupiec's performance test is performed afterwards separately. In the meantime, with the comparison and analysis of the test results on the VaRs gained by the above two methods with 95% and 99% confidence levels, the main conclusion is drawn: for the VaR method in the market risk measurement on the open-ended equity funds in China, it is a better way to choose one of GARCH related models(for different funds may be fit for different GARCH models) as the volatility model to describe the return's volatility clustering characteristic, to choose student T distribution hypothesis to capture the return's leptokurtosis characteristic. If a low confidence level is chosen, it is better to calculate VaR by Monte Carol simulation method; if a high confidence level, parameter method will comparatively be a better choice.