Research On The Value At Risk Based On Shanghai 180 Index

With the deepening of economic globalization degree and the rapid rise of the financial sector,the opening degree of the world financial market is constantly expanding in recent years,the connection between the global various countries increasingly close,which prompted capital throughout the world is more free to flow and quickly,linkage effect between countries is more and more obvious,financial markets are inextricably linked,one or more volatility in financial markets will quickly spread to other financial markets associated with them,again with the conduction,there would be a cascade of financial crises.Due to more liquidity and a higher degree of freedom,the financial sector regulation is facing many challenges.When vigilance is lacking and controls are not tight,financial markets become more volatile.In this case,financial risks will become more and more harmful and hidden.All kinds of financial institutions have to take on more risk,but also face a more severe and complex financial market environment.Once the risk occurs,both investors and financial institutions will suffer a huge impact,and sometimes even rise to the national level,leading to the country's financial crisis.Therefore,how to carry out effective risk measurement to ensure that measures can be taken to minimize the risk and reduce the corresponding losses of financial market and investors before the risk comes is of great significance to the country and investors.This paper mainly studies the risk quantification of stock.Firstly,the selected stock data is preprocessed,which includes data cleaning and data conversion.Data cleansing is the elimination of invalid values without affecting the total number of samples.Data conversion measures risk by using the definition of logarithmic rate of return,because it will reduce the impact of rounding error on the model.Secondly,it introduces the method of risk definition and quantification,among which VaR method is mainly introduced.Then,the method of calculating VaR is discussed in detail.In this paper,four calculation methods are mainly selected,including covariance matrix method,historical simulation method,GARCH model measurement method and extremum method.The theoretical basis and application environment of the four methods are introduced and compared.Based on the Shanghai 180 index,this paper uses the historical simulation method,the covariance matrix method,the GARCH model method and the extremum method to quantitatively analyze the risk value.Four methods are used to calculate VaR for data.Since it is difficult to judge the sequence distribution easily,the historical simulation method,which does not need to assume the distribution,can be used to calculate the range of VaR value.Then,using the method of variance covariance,the VaR of risk can be obtained by assuming that the variance is constant and the rate of return follows a normal distribution.The above two methods are simple to operate,but the calculation results are often quite different,so a more appropriate method is needed,namely GARCH model method.First,descriptive analysis was conducted on the stock data to obtain the time series diagram and basic statistics,and then the data characteristics were judged.Then,an appropriate GARCH model was established by using R,and the model with the highest fitting degree was selected by AIC and BIC criteria.Finally a kind of method is using extreme value idea,need to select threshold method,parameter estimation under different threshold and error,using the maximum likelihood method to estimate GDP model parameter values,and the final model validity check,if the inspection by,the good model may be used to calculate the VaR value of the stock data.Finally,it is necessary to carry out post-test,and use the likelihood ratio test statistics proposed by Kupiec to test the reliability of the VaR estimated value obtained under the four methods,and compare the results.Through the validity test of each model and the comparison of the estimated VaR value,the advantages and disadvantages of four different methods are analyzed and the most effective risk model is selected to provide a reference for financial institutions and investors.