Reaction to Common Outcome Shifts: An Exploration of the Connection Between Ellsberg and Allais Paradoxes
How to cite
Sung, Po Hyun; Sprenger, Charles; Adeney, Jack; Wang, Yu-Hsiang. 2024. "Reaction to Common Outcome Shifts: An Exploration of the Connection Between Ellsberg and Allais Paradoxes." AEA RCT Registry. AEARCTR-0014312. https://doi.org/10.1257/rct.14312-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
The Allais and Ellsberg paradoxes are two of the most widely studied problems in individual decision-making, highlighting violations of expected utility for objective risk and subjective ambiguity. One notable similarity between the two problems is that both involve an outcome shift between questions whereby the outcome on one state is shifted from strictly positive to zero. Motivated by the observation that these problems, while structurally similar, yield qualitatively distinct outcomes, we systematically use paired valuation tasks to investigate the space between the two. To pin down the source of deviations, we introduce three intermediate tasks that vary across three parameter domains- 1) different payoff schemes, 2) objective or subjective probability of the state of the outcome shift, and 3) objective or subjective probability of the remaining states. Finally, we vary the objective probabilities of the Common Consequence problem that range from the tuples that coincide with the Ellsberg problem and the classic Common Consequence problem.
Intervention
Each task assigned consists of a pair of valuation questions. Subjects will face ten tasks that are divided into two groups. The first group contains six tasks, and the second group contains the remaining four. Group 1 is always seen before group 2, but the order of tasks within each group is fully randomized.
Experimental design
This experiment elicits valuations in versions of the Ellsberg three-color and common consequence problems using multiple price lists. All subjects complete all valuation tasks. Design calls for comparisons both between subjects (first task) and within subjects (all tasks).
Design details
Cited by
No citation data available for this registration yet.