A Comparative Study Of Portfolio Strategies Based On Fuzzy Concepts

In recent years,the financial market has gradually occupied an important position in social economic activities.However,ordinary investors have strong blindness and herd mentality,and cannot make more correct investment choices.Therefore,it is of great significance to study how to combine assets effectively so as to get a better investment strategy.In this paper,a new measurement method of portfolio risk,the uncertainty of portfolio fuzzy returns,is proposed.In terms of data selection,this paper selects 40 constituent stocks of the SSE 300 Index as the research objects for the experiment in this paper.At the same time,it also comprehensively considers the situation of scale and industry differences when choosing stocks.Based on the famous Sharpe ratio in modern portfolio theory,this paper proposes the so-called fuzzy Sharpe ratio in the background of fuzzy modeling.In addition to introducing a new risk measurement method,this paper also proposes a return uncertainty ratio for evaluating portfolio performance in a fuzzy model.According to the TM fuzzy algorithm,based on the above proposed fuzzy sharpe ratio and earnings uncertainty index,two kinds of portfolio optimization model is established,the two models is to maximize the benefits of the objective of minimizing and risk,but when the build targets,differences between two models,the first model is under the condition of the introduction of fuzzy factors,converting the stock returns to fuzzy earnings,and according to the TM method to build the required to get returns and risk of expression,thus turn problem for multi-objective optimization problem,using genetic algorithm to solve the second model is not the introduction of fuzzy factors,Taking the average return rate of the selected stocks on January 1,2017 solstice on January 1,2020 as the return target of the portfolio and its standard deviation as the risk target,this model also converts the problem into a multi-objective optimization problem,which is also solved by genetic algorithm.This paper also makes innovations in the selection of data.It not only selects the historical stock data of China and the United States,but also takes into account the scale and industry differences in the selection of stocks.In addition,this paper also compares the difference of portfolio performance between using fuzzy numbers to measure returns and not using fuzzy numbers to measure returns.