Research On Dynamic Portfolio Based On Loss Aversion And Skewness

With the development of the domestic financial market,the financial system is becoming more and more perfect.The introduction of a large number of foreign quantitative models makes the portfolio model more abundant.How to choose an appropriate portfolio model to reflect investors' true investment psychology,and thus to make rational allocation of assets,is a hot issue in the industry.Therefore,in this context,this paper puts forward the improved correlation model in this paper,trying to put forward some guiding significance and practical significance for the application of China's stock market.Based on the prospect theory put forward by Kahneman & Tversky,the investment psychology of investors is depicted on the basis of the S-type utility function proposed by the traditional portfolio model.On the basis of previous research results,the model is improved,and the dynamic loss aversion portfolio model is established by changing the loss aversion coefficient and reference point in different periods.On the basis of the improved model,three order moment skewness is introduced to study the influence of higher order moments on the model.Finally,in order to verify the robustness of the model,the transaction cost is introduced into the model,and the validity of the model is proved by empirical analysis.In the process of solving the model,due to the inconvenience caused by the segmentation of parameters,this paper simplifies the model by proposing a reasonable hypothesis and introducing 0-1 variables.The NSGA II algorithm is used to solve the efficient frontier of multiobjective programming.The algorithm is an optimized non inferiority sorting genetic algorithm,which is an evolutionary version of NSGA,and has been improved in terms of algorithm complexity,elitist strategy,and so on.By taking A stock data for empirical research,this paper draws the following conclusions.The improved dynamic loss aversion model is superior to the improved model and the static portfolio model.The performance of the dynamic loss aversion portfolio model with skewness is better than the dynamic loss aversion portfolio model without skewness.After introducing transaction cost,it is proved that the improved dynamic loss aversion portfolio model is effective,and the robustness of the model is also verified.