A cloud theory-based multi-objective portfolio selection model with variable risk appetite. (15th August 2021)

• Record Type: Journal Article
• Authors: Gong, Xiaomin; Yu, Changrui; Min, Liangyu
• Is Part Of: Expert systems with applications. Volume 176(2021)
• Publication Date: 2021-08-15
• Subjects: Portfolio selection -- Multi-objective -- Cloud model theory -- Risk appetite -- Compromise programming; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2021.114911
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: A cloud theory-based multi-objective portfolio selection model is proposed. The crisp numerical characteristics of the cloud model theory are defined. The acceptance and rejection functions with variable risk appetite are constructed. A compromise programming approach is utilized to aggregate the objectives. Several schemes as per investors' preferences regarding all objectives are devised. Abstract: This study proposes a cloud theory-based multi-objective portfolio selection model with variable risk appetite, which incorporates four objectives of mean, variance, skewness, and liquidity constrained by several realistic constraints. Cloud model theory is employed to characterize the return rates and liquidity of assets due to the superiority of simultaneously capturing the ambiguity and randomness of information. The crisp numerical characteristics of the cloud model are defined to obtain the crisp form of the proposed model. To highlight and portray the investors' risk (averse-neutral-seeking) appetites, the generalized acceptance and rejection functions are modeled by using the extreme values of each objective and introducing a variable risk appetite parameter. Thus the corresponding model is transformed with the objective functions of maximizing acceptance and minimizing rejection, which is solved through the compromise programming approach. The extended model provides investors with an opportunity to adjust risk parameters according to current market status. Moreover, the preference ratio vector is introduced when optimizing, which provides investors with overall control over the preferences regarding all objectives, so that investors can derive optimal portfolios well compatible with their expectations through customized weighting schemes. A real-world empirical application is presented to demonstrate the effectiveness of the proposed model