A Study Of Portfolio Optimization Model Based On CVaR And Empirical Comparison Research

Methods and theories, used to research on risk measurement and optimization of securities portfolio, have significant meaning as to finance risk control and finance order stabilization. At first, the research in this paper is based on the framework of Markowitz portfolio optimization theory. In this framework, this paper briefly introduces theoretical foundation of modern portfolio in returns, risk, and strategy aspects. Then, this paper stresses the measurement of risk, an important and difficult part on research, systematically analyzes three basic methods of risk measurement of VaR, and compares with their own merits and dismerits. Uryasev and Rockafellar put forward CVaR to improve VaR because VaR is not a coherent risk measurement method and its arithmetic is complex. This paper further analyzes the relationship between VaR and CVaR and discusses linear programming and extreme theory,two kinds of measurement method in CVaR.: This paper researches on portfolio Optimization model based on VaR and CVaR, uses geometric method to compute Mean-Variance Model with VaR constraint, and constructs Mean-VaR Model and Mean-CVaR Model that can be transformed into linear programming.At last, by using Matlab and Excel, this paper makes empirical comparison to innovatively research on portfolio and efficient frontier of Mean-Variance Model, Mean-VaR Model, and Mean-CVaR model, and then shows that Mean-CVaR Model not only share better theoretical traits but also is more suitable for securities market in China and is more practically meaningful. During the research, this paper put forward to combines simple returns and log returns and use Monte Carlo simulation method to simulate and anticipate rate of returns of portfolio. And then, this paper systematically researches on the influence of constraint condition, such as confidence level, trade cost, and quota system of lowest capital, on Mean-CVaR portfolio optimized model.This study is sponsored by the National Natural Science Foundation of China (Grant No. 70353076) and Research Foundation of Doctoral Program of higher Education(Grant No. 20050056057).