互联网金融资产的多目标投资组合研究

互联网金融的快速发展和不断创新,正在悄然改变着公众的投资理财行为。本文在分析互联网金融创新下公众流动性偏好、投资行为变化与资产选择的基础上,构建基于CRRA(常数相对风险厌恶)期末财富期望效用最大化和VaR最小化的多目标投资组合模型。同时引入多目标优化的NSGA-Ⅱ遗传算法,并选择实际数据对模型进行求解,得出最优的互联网金融资产组合。研究表明:(1)互联网金融给传统金融业带来冲击的同时,也改变了人们的流动性偏好、投资行为和资产组合选择。(2)互联网金融在一定程度上调和了金融资产“流动性、收益性和安全性”之间的矛盾,并兼顾了“三性”的相对统一。(3)模型求解结果显示,投资者对互联网金融资产的投资组合为低风险类资产60%左右、高风险类资产40%左右。

A Study on the Multi-Objective Portfolios of Internet Financial Assets

In recent years, the rapid development and continuous innovation of Internet finance in China have greatly affected the traditional financial industry and portfolio theory. Against such a backdrop, it is a relatively new and important issue to study the changes in public liquidity preferences, investment behavior, and asset selection. However, the research is relatively lacking and has little connection with portfolio theory. This paper analyzes liquidity preference, investment behavior change, and asset selection under Internet financial innovation to reveal the influencing mechanism of Internet finance on public investment behavior. It also attempts to build a multi-objective portfolio model based on the CRRA’s expected utility maximization and VaR minimization at the end of the period. Finally, this paper introduces the multi-objective optimization NSGA-Ⅱ genetic algorithm to solve the model. It determines the optimal portfolio weights and proposes the corresponding countermeasures and suggestions.
This paper constructs a multi-objective portfolio selection model of Internet financial assets based on the maximization of the expected utility of the CRRA and the minimization of the VaR. Furthermore, it introduces a multi-objective optimization NSGA-II genetic algorithm to solve the model. To overcome the subjectivity of the model parameter setting, this paper calculates the return rates of various assets on the basis of calculating the actual data of various assets and combining the characteristics of Internet financial products. The data come from the Wind database and Internet Loan Home. The simulation results show that in the portfolio of public Internet financial assets, the weight of low-risk financial assets is 59.62%, of which 27.71% are Internet Monetary Fund products and 31.91% are Internet insurance products. The weight of high-risk Internet financial assets is 40.38%, of which 27.91% are crowdsourcing products and 12.47% are P2P products. This shows that low-risk Internet financial products are popular with public investors, whereas high-risk products meet the investment needs of some investors who pursue high yield with high yield and convenience.
This paper suggests that the public should follow the principle of optimizing platform, diversifying investment, pursuing income, and ensuring safety in the investment decision-making process. Internet financial enterprises or financial institutions should fully reflect three relatively unified characteristics when developing Internet financial products. Effort should be made to give the developed products the advantages of high yield, high liquidity, and low risk and to better provide personalized financial services for the general public. The regulatory authorities should establish and improve relevant laws and regulations, standardize the development of Internet finance, create a good market environment, promote innovation of Internet financial products, meet diversified investment needs, and safeguard the rights and interests of investors.
This paper makes three main contributions.First, it proposes that Internet finance accounts for the relative unity of the liquidity, profitability, and security of financial assets and carries out a theoretical analysis and numerical simulation. Second, on the basis of comparing traditional financial products with Internet financial products, this paper constructs a multi-objective portfolio model based on the CRRA to maximize the expected utility of wealth at the end of the term and to minimize the VaR, which can be used to analyze the public portfolio behavior under Internet financial innovation. Third, it introduces the multi-objective optimization NSGA-II genetic algorithm to solve the multi-objective model. The actual yield data of various Internet financial products are used to simulate the model and the optimal portfolio is obtained, which can serve as a reference for public investors.