Research On Portfolio Selection Model Considering Liquidity Constraints Based On Credibility Theory

Since the birth of Markowitz’s classic mean-variance model,scholars have carried out many explorations on portfolio selection models and have achieved fruitful results.Taking into account the uncertainty of real financial market information,scholars began to study the portfolio selection model under the framework of fuzzy theory,and expanded the model based on the possibility theory.Later,in order to solve the non-self-dual defect of the possibility theory,scholars further developed the credibility theory,which provided an axiom basis for the fuzzy theory.In addition,the traditional portfolio selection model does not consider the liquidity of securities,which may lead to liquidity risks.The advantage of the turnover rate indicator is that it can capture the monthly changes in the liquidity of securities and check the liquidity level of a large number of securities at the same time.Therefore,based on the credibility theory,the turnover rate is used to measure the liquidity level of securities and research is conducted to analyze the multi-objective portfolio selection model considering liquidity constraints in a fuzzy environment.Firstly,based on the credibility theory and the mean-variance-skewness model,the turnover rate index is introduced into the model as a constraint to measure the liquidity of risky assets in order to avoid liquidity risks caused by being unable to sell for a long time or having to sell at a lower price.At the same time,in order to make the model more suitable for investors’ actual psychological definition of risk,semi-variance is used as a risk measurement index to construct a mean-semi-variance-skewness single-stage portfolio selection model considering liquidity constraints.The historical data from the Shanghai Stock Exchange and the NSGA-III algorithm is used to simulate the real investment environment and perform empirical analysis.By pre-setting the standard value of the turnover rate,a portfolio of securities that meets the investor’s liquidity preference is generated to verify the flexibility of the model.Secondly,taking into account that in actual situations,investors generally conduct continuous multi-stage investment behaviors rather than single-stage behaviors,further expanding the model to a multistage portfolio selection model;taking into account that transaction costs occur at various stages of asset purchases,transaction costs are introduced into the model,and a mean-semi-absolute deviation-skew portfolio selection model considering liquidity constraints and transaction costs is constructed.The multi-objective artificial bee colony algorithm and real historical data is combined to solve the problem,and proves that the model can reasonably adjust the investment ratio of each asset at the beginning of each stage according to the change of asset return rate index and turnover rate index,verifying the effectiveness of the model.Finally,in order to further prove the advantages of considering liquidity constraints in the portfolio selection model,the constructed models were compared with the corresponding models that did not consider liquidity constraints.The empirical results prove that,when the historical transaction data and turnover rate data are kept consistent,compared with the model that does not consider the liquidity constraint,the model that considers the liquidity constraint is more rigorous in the selection of the optimal solution and can comprehensively consider the factors of return and liquidity,so that the generated results can better ensure that investors can obtain the maximum return under the liquidity risk that they can bear,which is more stable in solving practical problems and has greater practical significance.