The Evaluation And Test Of Sharpe Ratioa Based On VaR And ES Adjusment In Funds

Securities investment funds are a key part of modern finance.Funds of varioustitles such as stock,bond,income and balance have started to flood the market,giving investors more choices and raising some concerns,for example,how to choose what you want to fund in so much of the fund,the fund held by the investment would be able to obtain continuous fund income etc.Therefore,it is very important for the fund investor to choose the scientific and effective fund performance evaluation index to evaluate the fund performance.However,a review of the evaluation of fund performance at home and abroad has found that the standard deviation of traditional Sharpe Ratio is the standard deviation of the yield.In empirical research evaluation is more focused on the fund risk rather than the entire risk as to loss,this is the standard deviation is unable to accurately depict the gains,because it ignores the most important risk of loss,namely the Downside Risk.In view of the disadvantages of the traditional Sharpe Ratio,the standard deviation of the return on investment is replaced by the index,the Value at Risk and the Expected Shortfall,which is used to measure the downside risk,thus adjusting the traditional Sharpe Ratio.First,the GARCH model is used to characterize the clustering of fluctuations in the time series of return on investment.Secondly,the extreme value theory is used to model the impact of extreme events,and the POT model is constructed to study the tail risk characteristics of the sequence.Finally,considering the thick tail characteristics and volatility clustering phenomena of the yield series,the GARCH-EVT model is improved by improving the GARCH model and the POT model to improve the validity of the risk measurement model.This paper selects 24 fund data from October 25,2006 to October 25,2016 for empirical research,and tests the effect of VaR and ES risk measure given by the three models metioned by Kupiec likelihood ratio test and Bootstrap method to measure the accuracy of the model,on the basis of theoretical research.The results show that the extreme value theory can better characterize the tail shape of the distribution,and improve the accuracy and validity of the model under the higher confidence level.The VaR and ES values estimated based on the GARCH-EVT model are more efficient than the other models,which means that the GARCH model and the extreme value theory can be combined to achieve the effective measure of extreme risk.The revised Sharpe ratio is more reasonable than the traditional one,in which the Sharpe ratio based on ES is more conservative than the Sharpe ratio based on the VaR correction.This thesis focuses on the estimation of Sharpe Ratio,which is different from other literatures.It studies the estimation of the traditional Sharpe Ratio and the estimation of the Sharpe Ratio based on VaR and ES under the actual daily rate of return of the fund.The hypothesis test is proved to be significant by using the statistical method of double sample hypothesis test.The conclusion proves that the hypothesis test is significant,and the application of SR in the performance evaluation of open-end fund in China is feasible.