Integrating prospect theory with variable reference point into the conversion-based framework for linear ordinal ranking aggregation

Nana Liu; Zeshui Xu; Hangyao Wu; Peijia Ren

doi:10.1007/s00500-022-07339-7

Integrating prospect theory with variable reference point into the conversion-based framework for linear ordinal ranking aggregation

Soft comput. 2022;26(21):11713-11732. doi: 10.1007/s00500-022-07339-7. Epub 2022 Aug 26.

Authors

Nana Liu¹, Zeshui Xu², Hangyao Wu³, Peijia Ren⁴

Affiliations

¹ School of Business Administration, Chongqing Technology and Business University, Chongqing, 400067 China.
² Business School, Sichuan University, Chengdu, 610064 Sichuan China.
³ School of Management Science and Engineering, Chongqing Technology and Business University, Chongqing, 400067 China.
⁴ School of Business Administration, South China University of Technology, Guangzhou, 510641 China.

Abstract

Considering that converting linear ordinal ranking (LOR) information into interval utility values can not only improve the computability of LOR information but also explore the degree of preference for different alternatives for decision-makers hidden behind LOR information, this paper proposes a conversion-based LOR aggregation method to aggregate LOR information under risk. Given that the behaviours of decision-makers are influenced by risk, this paper adopts prospect theory to depict the decision-makers' behaviours under risk in the conversion-based aggregation process. To achieve this, the information energy for LOR is constructed firstly, and its features are analysed, which makes a basis for the conversion process. After that, the details about how to integrate the prospect theory with variable reference points into the conversion-based aggregation framework are presented. Finally, an example (exploring the financial product preferences of a group of respondents) evidences the practicality of the proposed method. Further, some analyses and discussions are conducted to verify the rationality and stability of the method.

Keywords: Information energy; Linear ordinal ranking aggregation; Prospect theory.