Fuzzy Multi-objective Portfolio Model With Psychology Of Investors And Interactive Algorithm

Markowitz mean-variance portfolio model created the beginning of portfolio theory which made modern financial investment theory from the stage of descriptive analysis into the stage of quantitative analysis.It provided powerful reference for investors'investment decision.Economic globalization and integration makes the complex securities market more volatile,and investors are difficult to make a decision and seek the optimal portfolio in the complex market environment.Therefore,the research of the portfolio optimization problem has important theoretical and practical significance.Considering the uncertainty of security market and the characteristics of different risk preference,a fuzzy multi-objectives portfolio investment model with risk attitude of inventors is built.The model regards the asset return and turnover rate as fuzzy variables,using the maximum investment return and liquidity,the minimum risk to be investment objectives,and the single asset investment proportion to be constraint conditions.In general,we use the interactive algorithm to solve multi-objective portfolio problems.In order to save computing time and allow investors to understand the intuitive sense of decision-making basis with the purpose of decreasing the decision-making difficulty,a Cumulative Prospect Theory and TOPSIS method based interactive algorithm is presented to solve the proposed portfolio model,where the interactive algorithm features investors' characteristics such as risk appetite,loss aversion and reference dependence.Finally,under the economic environment of domestic stock market,the effectiveness of the model and its algorithm is empirical study using the constituent stock of Shanghai 50 index as the research sample.The research mainly includes that comparing the structure and the target value of portfolio which different risk attitude investors choose and testing the investment performance of model and algorithm through the comparison with accumulative return rate of Shanghai 50 index.The result shows that the modified algorithm and the model not only have a better outcome than security market,but also reflect investors' psychological characteristics which include risk preference,loss aversion and reference dependence,which verify the validity of the model and algorithm and can be the dependent of investor's real decision.