Research Of The Mean-CVaR Portfolio Model Based On Transaction Cost

The stock market is a high risk market,it is difficult to predict the market trend because of many uncertainties,investors are constantly looking for investment strategies that can effectively reduce the risk of investment and get higher returns。A large number of studies have shown that a reasonable and effective combination of investors can meet the needs of investors.People have been studying various methods to expand and improve the securities selection model since Markowitz proposed mean-variance model,so that the model can be more suitable for the market conditions or better measure the risk,to better serve the investment practice.In order to establish a more practical securities selection model with market conditions.this paper introduces the transaction cost on the basis of the traditional mean-CVaR portfolio model.Firstly,this paper summarizes the relevant research results at home and abroad,and describes the basic theory of portfolio,CVaR and genetic algorithm,in addition we use a variety of market index to verify the asset return rate of China’s securities market subject to the Laplace distribution,and illustrate the assumptions of the model construction,on this basis,the mean-CVaR portfolio model based on transaction cost is established.Then according to the historical simulation method to estimate the expected rate of return,the SSE 50 Index constituent stocks to construct initial optimal securities portfolio.At last,taking an active strategy to adjust portfolio according to the optimal securities portfolio which is calculated by the model,adjusting once a month frequency is three times,at the same time,the return of portfolio investment in each period is calculated,and the return is compared with the consolidated income of the SSE 50 Index and portfolio returns without considering transaction costs.The empirical results show that the positive strategy of the mean-CVaR portfolio model based on transaction cost is effective.