Research On Cooperative Decision Model Of Investment Preference Based On Primary And Secondary Impact Factors

With the increasing marketization of the financial industry,wealth management has become an indispensable part of people's life.The common concern of investors is how to get the highest possible return while keeping the investment risk low.Portfolio optimization can disperse risks by means of asset allocation.However,as this problem is essentially a multi-objective optimization problem,most of the current researches can only obtain a group of optimal solutions,while investors usually only need satisfactory solutions that conform to their own preference characteristics.Therefore,it is of great theoretical significance and application value to introduce investors' preference characteristics in the solution method to help investors find their satisfactory solutions.A common method for introducing preference features of decision makers is an interactive evolutionary multi-objective optimization method,which contains three links,namely evolutionary algorithm,preference feedback and preference model.In the existing researches,investors usually only rely on the value of the target to be optimized when giving preference feedback,and the form of preference model is usually a polynomial function in a fixed form.This paper improves the method in the aspects of preference feedback and preference model,and designs a collaborative decision-making model of investment preference based on the main and auxiliary impact factors.In the research on the collaborative preference model of investment preference based on primary and secondary factors,a preference feedback model based on main and auxiliary impact factors was first designed.This model enables decision makers to be assisted in decision-making based on other related information(auxiliary factors)in addition to the return and risk values(main factors)in the process of preference feedback,so as to avoid investors from making uncertain decisions due to the extreme similarity of primary factors.Therefore,it is necessary to analyze the factors that may affect the investor's decision and determine the candidate secondary impact factors.In combination with the investor's psychology of mainly caring about returns and risks,this paper also considers the correlation between auxiliary factors and main factors and the differentiation degree of auxiliary factors themselves,and designs an auxiliary factor selection mechanism based on correlation degree and supplemented by differentiation degree.Secondly,in order to ensure the objectivity of the experiment,this paper makes assumptions about the decision makers' preferences,uses nonlinear functions to simulate the decision makers' preference judgment,and designs the decision makers' judgment process based on the main and auxiliary factors.Finally,because the decision tree can learn the investor's preference structure through preference feedback and can respond flexibly to preference changes,this paper uses the decision tree as the preference model.The practical application of the collaborative decision-making model of investment preference shows that the model can learn the preference of decision makers flexibly and accurately,so as to effectively help decision makers to find their satisfactory portfolio plans.In addition,the validity of the collaborative factors and preference model was tested in this paper,which verified that for decision-makers of any risk aversion type,no matter at what rational level,the existence of collaborative factors could make the preference feedback result more accurately reflect their preference tendency,so as to reduce the feedback noise.It is proved that the preference model based on decision tree can describe the preference information of decision maker more accurately than the general polynomial function.