Research On Portfolio Risk Measurement Of Open-end Fund

With the increasingly close global economic and trade relations,the correlation between different financial assets is becoming more and more complex,the impact of uncertain events on the financial system is increasing,and portfolio risk management is becoming more and more important.The traditional linear correlation analysis method can not accurately reflect the correlation between financial assets.At the same time,the financial time series has obvious characteristics of peak and heavy tail,showing the non-normal distribution and non-linear correlation.Therefore,the traditional risk value measurement method seems unreasonable.At this time,it is necessary to introduce a new measurement method.Copula function,as a kind of connection function,can better deal with the complex correlation between financial assets,better describe the distribution of financial assets return rate,and then establish a more reasonable risk measurement model.This paper is divided into five parts.Chapter 1 is an introduction.It mainly introduces the research background and significance of this paper,and combs the relevant literature.Chapter 2 introduces the theoretical basis of the model,including the definition of GARCH function family,Copula function and extremum theory.Chapter 3 introduces the research methods of VaR,and constructs the Copula-GARCH model,which makes preparations for the empirical research of Chapter 4.Chapter 4 is an empirical study.This paper uses four open-end funds as sample data to analyze the correlation between multi-dimensional variables,and through Monte Carlo simulation and historical data for comparative analysis.Chapter 5,on the basis of empirical research,analyses the research results of this paper,and gives countermeasures and suggestions based on the problems existing in the open-end fund market in China.In this paper,the sample data are processed preliminarily before the Copula estimation of the four funds.Firstly,the AR(1)-GJR(1,1)-t model is used to filter the sample data and obtain the standard residuals.Then the tail of the standard residuals is fitted by GPD and transformed into probability density to obtain the uniform distribution on [0,1].Finally,t-Copula is used to estimate the correlation of four variable.The results show that Copula-GARCH-GPD model can better characterize the dependence structure and tail characteristics of portfolio.The results show that Copula-GARCH-GPD model can better characterize the dependence structure and tail characteristics of portfolio.The range of fund volatility is closely related to the types of funds.Open-end funds with a large share of stocks have obvious correlation.Adding bond funds into portfolio can reduce the risk of portfolio.