Research On Fuzzy Portfolio Strategy Of Emotion Cognition Based On Stock Price Forecasting

The high returns brought by the financial market make people focus on this,and investors have invested their assets in order to realize the appreciation of assets.Investors need to go through two stages when making stock investment.The first stage is to predict the future development trend of stocks and select better stocks for investment.The second stage is to choose a portfolio.Investing in a single asset will make investors take greater risks.Investors often choose portfolio investment to diversify their risks.Aiming at the two stages of investment that the above-mentioned investors need to face,this paper uses the Support Vector Machine(SVM)theory,Markowitz mean-variance theory and fuzzy possibility theory to construct an emotional cognitive fuzzy investment portfolio based on Bayesian optimized Support Vector regression stock price prediction.The main research contents are as follows:In the process of stock selection,investors choose stocks mainly based on the expected return of stocks,and the calculation of expected return is based on the prediction of future stock prices.Therefore,in order to obtain the expected return of stocks,it is necessary to Predict the price.This paper uses the Support Vector Regression(SVR)model as the basic model of stock price prediction,and compares and analyzes the characteristics of different kernel functions and model prediction effects.The Gaussian radial basis kernel function is selected as the kernel function of SVR,and the necessity of parameter optimization is discussed on this basis.Using Bayesian Optimization Algorithm(BOA)to optimize the parameters C,? and ? of SVR,a BOA-SVR stock price prediction model is constructed.Comparing the efficiency and effect of Bayesian optimization algorithm,grid search algorithm and random search algorithm to optimize SVR,it is verified that BOA-SVR model has better performance and prediction effect than other optimization algorithms to optimize SVR model.The second stage of investment is to construct an effective investment portfolio and rationally allocate investors' funds.This article conducts research from two aspects: the ambiguity of financial markets and investor sentiment.Since investors' satisfaction with securities investment income,securities liquidity and overall market situation cannot be described with certain numerical values,the financial market is not only random,but also vague.In this paper,a fuzzy portfolio model with expected return rate of fuzzy number is constructed based on the fuzzy possibility theory under the fuzzy uncertain environment.Moreover,in the process of investment,investors' psychology and behavior and other factors will have an important impact on the portfolio strategy.This paper introduces the principle of dominance of investment sentiment,integrates investor's emotional cognitive return and variance of emotional cognition,and constructs a fuzzy portfolio model considering emotional cognition.