Research On Sine Entropy Porffolio Optimization Model Based On Credibility Theory

Since Markowitz proposed the mean-variance(MV)model,portfolio theory and practice have developed rapidly.However,the risk measure in the MV model is variance,and there are some obvious defects,such as the assumption that the conditions do not meet the actual situation,high yield penalty,parameter estimation instability,covariance matrix complexity,etc.,these factors greatly affect its risk Measure the persuasiveness of empirical research.In recent years,scholars have applied it to the field of portfolio theory based on the excellent nature of information entropy,and put forward a variety of entropy index risk measures such as fuzzy entropy,mixed entropy and cross entropy.Based on the previous research,this paper chooses the sine entropy as the securities risk measure to study the portfolio model based on the sinusoidal entropy risk measure in the fuzzy stochastic uncertainty environment,and add the fuzzy flow constraint to optimize the model,Study the effect of the test model.The main work is as follows:(1)Using Wind information to collect the daily trading data of fifteen stocks of different industries in Shanghai 2014-2015,and using the Markov forecasting method and K-means clustering method to preprocess the sample data.The stock securities triangular fuzzy yield and gradient fuzzy liquidity index data.Lay the foundation for the follow up study of the paper.(2)M-V-S-SE and multi-objective Mult-M-V-S-SE based on the reliability mean-variance-skew-sine entropy are established by using the"credibility" theory as fuzzy membership degree measurement method.The M-V-S-SE model is superior to the Mult-M-V-S-SE model and can adjust the income and risk forecast according to the investor’s expectation to meet the different requirements,be more stable and flexible.(3)Considering the liquidity factor in the real stock market,the fuzzy liquidity constraints is added in the model,and the mean-variance-skew-sine entropy investment portfolio optimization model,and then improve the empirical effect of the model.Through the comparative empirical study,it is found that the improved model has the highest cumulative yield and the stability of the model is further enhanced,which can avoid the investment risk to a certain extent.With the development of diversified capitalization and margin trading business,the involvement of venture capitalists makes investors and regulators face severe risk challenge.By establishing mathematical model and optimizing the model to solve the problem of financial market securities investment.The results of combination analysis and empirical analysis show that the investment risk can be obtained by continuously adjusting the investors acceptable earnings,risk and liquidity expectation to avoid the investment risk in a timely and effective way.