The Improvement Of Conditional Value-At-Risk Portfolio Model Research

With the fast development of the society and the economy, the economic cooperation between countries and regions is increasingly getting closer. The trend of globalization and integration between the economy and the financial market on the one hand brings a lot of chance for the development of the countries and the regions; but on the other hand, it also has a huge impact on the financial market. This trend arouses wide concerns in that it greatly influences the stability of the financial market and increases the risk of investment. In order to lower the risk and gain reasonable expected returns at the same time, many investors disperse their assets into several parts to form an asset portfolio. Now investment portfolio has been an important issue in financial research. In 1952, the famous economist Markowitz brought people into the field of financial qualitative analysis. It is a successful finding, but later, people find that using the model of mean and variance to measure the portfolio investment is deficient, and this model is hard to operate in reality. Since then, more researches have been done in order to overcome this problem. But it still needs to make improvements in the measurement method and the model of portfolio. Therefore based on the mean-variance model and the mean-CVaR model, this current study establishes a multi-objective portfolio model of CVaR containing asset management fees,and a SCVaR portfolio model containing an optimized asset management cost.Firstly, this thesis will overview the terms of mean-variance, mean-semivariance、CAPM、mean-VaR、mean-CVaR model. In order to get a full and real picture of the investment portfolio market, it is also essential to research the investment environment, to consider the management fees and to appropriately use various kinds of risk measurement.Secondly, this thesis will also study the degree of difficulties in managing the assets by firstly defining the cost function of asset management, then building up the multi-objective portfolio model based on the measurement of CVaR. Consider the advantages of CVaR, by introducing the uncertainty of entropy into the investment portfolio model, a newly improved investment portfolio model is built up.Finally, by using the intelligence algorithm to solve this model, this thesis will compare and verify the feasibility, validity and superiority of this model, and thus offer a new investment portfolio for the investors.