Research On The Construction Of Stock Portfolio Based On Multi-objective Water Cycle Algorithm

The stock market is an indispensable part of China's market economy,and with the continuous improvement of China's comprehensive national strength,people's disposable income is also increasing,so more and more people choose to purchase stocks for financial management.Therefore,how to reduce the risk of stock investment under the premise of ensuring return has become the focus of attention,and this is also the key problem to be solved by portfolio model.Portfolio theory describes the stock investment problem as how to maximize the investment utility under uncertainty,that is,to pursue the optimization of the ratio of return and risk.However,most of the traditional portfolio models only take the closing price of the stock as the analysis variable,which only includes the technical analysis of the rise and fall of the stock,and ignores the importance of the financial information of the listed companies and the importance of the basic analysis,which will have a negative impact on the effectiveness of the portfolio.Therefore,in order to overcome this defect,this paper combines the historical information data of stocks,financial information data of listed companies and the model of multi-objective optimization algorithm to study and analyze the construction of portfolio.First of all,because there are many kinds of stocks in China's stock market,the quality of stocks is also uneven,so this paper uses the sentiment intention index(BRAR)to screen the appropriate stocks in China's A-share market,and selects a total of 17 stocks in A,B and C groups;then,with the maximum return and minimum risk as the objectives,the Multi-objective Water Cycle Algorithm(MOWCA)is used.The paper constructs the stock portfolio without weighted upper bound and the stock portfolio with weighted upper bound respectively;then,taking the maximum yield,minimum risk and maximum default distance as the objectives,MOWCA is used to construct the stock investment portfolio without weighted upper bound and the stock portfolio with weighted upper bound respectively;finally,the stock investment portfolio proposed in this paper is used in the portfolio construction model disperses the risk of heavy position stock,which makes the investment portfolio not only obtain the same income as the heavy position stock,but also greatly reduces the risk of stock investment.The research results show that:(1)for the selection of stock portfolio,because the systematic risk of stock investment is inevitable,we can only reduce the non-system risk of stock investment as far as possible,so everyone can choose on the frontier of stock investment portfolio according to their own risk tolerance ability,or according to their expected return rate.(2)In general,the total return of the stock portfolio is proportional to the risk,that is,with the increase of the total return,the risk increases simultaneously.The higher the yield of the stock portfolio,the greater the default distance,the less likely it is to default.However,there may be special situations.(3)Combined with default distance,yield and risk,this paper constructs the stock portfolio.First of all,the default distance should be considered.It is necessary to eliminate the portfolio whose default distance is less than 1.96(the probability of default is greater than 5%),so as to avoid the influence of internal problems of listed companies on the return of stock portfolio.Then,we should choose the investment weight of stock according to the risk tolerance of stock buyers,and choose the portfolio with higher yield and larger default distance within the risk tolerance range.(4)Setting the upper limit of the weight of each stock in the stock portfolio can avoid the unreasonable extreme stock weight and affect the analysis of the stock portfolio weight.It can also increase the diversity of the stocks in the portfolio and effectively disperse the stock investment risk.(5)The empirical results show that the average return of group A is the highest among the three groups,the risk of group B is the lowest,and the volatility of return of group C is the largest among the three groups,and the value of risk is higher.Therefore,the investors who pursue stability can find the weight ratio of investment from the stock portfolio of group B,and the investor pursuing high return can find the weight ratio of investment from the stock portfolio of group A.The stock portfolio construction model proposed in this paper broadens the related research on portfolio construction.At present,most of the construction of the stock portfolio only uses the closing price of the stock.This paper combines the default distance data of the listed companies calculated by KMV model and the historical closing price data of the stocks,and brings the financial data of the listed companies into the construction of the portfolio model,so as to realize the construction of the portfolio model from a new perspective.In addition,the empirical study shows that the model can also be used to spread the risk of heavy positions of stocks,and can achieve a better effect of ensuring return and risk dispersion.To sum up,the model proposed in this paper can determine the stock portfolio according to the internal financial information and stock history information of listed companies,improve the effect and stability of the investment portfolio,effectively avoid the non systematic risk of stock investment,and play a certain guiding role in the actual stock investment.