Affiliations: [a]
School of Statistics and Data Science, Qufu Normal University, Qufu, China
| [b]
School of Mathematics and Computer Applied Technology, Jining University, Qufu, China
Correspondence:
[*]
Corresponding author. Chuancun Yin, School of Statistics and Data Science, Qufu Normal University, Qufu 273165, China. E-mail: [email protected].
Abstract: Probabilistic linguistic term sets (PLTSs) provide a flexible tool to express linguistic preferences, and several multi-criteria decision models based on PLTSs have been recently developed. In this framework, distortion risk measures are extensively used in finance and insurance applications, but are rarely applied in fuzzy systems. In this paper, distortion risk measures are applied to fuzzy tail decisions. In particular, three tail risk measurement methods are put forward, referred to as probabilistic linguistic VaR (PLVaR), expected probability linguistic VaR (EPLVaR), and Wang tail risk measure and extensively study their properties. Our novel methods help to clarify the connections between distortion risk measure and fuzzy tail decision-making. In particular, the Wang tail risk measure is characterized by consistency and stability of decision results. The criteria and expert weights are unknown or only partially known during the decision making process, and the maximising PLTSs deviations are showed how to determine them. The theoretical results are showcased on an optimal stock fund selection problem, where the three tail risk measures are compared and analyzed.
Keywords: Probabilistic linguistic term sets, probabilistic linguistic VaR, expected probability linguistic VaR, Wang tail risk measure, maximizing deviation method
