Source Theory: A Tractable and Positive Ambiguity Theory

• 成果类型: Article
• 署名作者: Baillon, Aurelien; Bleichrodt, Han; Li, Chen; Wakker, Peter P.
• 署名单位: Centre National de la Recherche Scientifique (CNRS); Universite Lyon 2; Ecole Normale Superieure de Lyon (ENS de LYON); Universite Jean Monnet; Universite Lyon 1; emlyon business school; Universitat d'Alacant; Erasmus University Rotterdam; Erasmus University Rotterdam - Excl Erasmus MC
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.2023.03307
• 发表日期: 2025
• 关键词: Subjective beliefs; ambiguity aversion; Ellsberg paradox; source of uncertainty

摘要

This paper introduces source theory, a new theory for decision under ambiguity (unknown probabilities). It shows how Savage's subjective probabilities, with sourcedependent nonlinear weighting functions, can model Ellsberg's ambiguity. It can do so in Savage's framework of state-contingent assets, permits nonexpected utility for risk, and avoids multistage complications. It is tractable, shows ambiguity attitudes through simple graphs, is empirically realistic, and can be used prescriptively. We provide a new tool to analyze weighting functions: pmatchers. They give Arrow-Pratt-like transformations but operate within rather than outside functions. We further show that ambiguity perception and inverse S probability weighting, seemingly unrelated concepts, are two sides of the same insensitivity coin.