Eliciting Subjective Probabilities Through Exchangeable Events: An Advantage and a Limitation

Authors:
Aurélien Baillon
Affiliations:
Department of Applied Economics, Erasmus University, NL-3000 DR Rotterdam, The Netherlands
Venue:
Decision Analysis
Year:
2008

Citing 0
Cited 8

From the Editors...

Decision Analysis
From the Editors---Deterrence, Multiattribute Utility, and Probability and Bayes' Updating

Decision Analysis
Using Bayes' Rule to Update an Event's Probabilities Based on the Outcomes of Partially Similar Events

Decision Analysis
From the Editor---Multiattribute and Intertemporal Preferences, Probability, and Stochastic Processes: Models and Assessment

Decision Analysis
From the Editors---Copulas, Group Preferences, Multilevel Defenders, Sharing Rewards, and Communicating Analytics

Decision Analysis
From the Editors---Brainstorming, Multiplicative Utilities, Partial Information on Probabilities or Outcomes, and Regulatory Focus

Decision Analysis
From the Editors: Probability Approximations, Anti-Terrorism Strategy, and Bull's-Eye Display for Performance Feedback

Decision Analysis
From the Editors: Group Decisions, Preference Elicitation, Experienced Utility, Survival Probabilities, and Portfolio Value of Information

Decision Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Two events are exchangeable for an agent when she is indifferent to permutations of their outcomes. Such events are thus revealed to be equally likely. If they are complementary, the subjective probability associated with each event should be 1/2 (assuming the additivity of the probability measure). This paper reports the results of an experiment that elicits probabilities through exchangeable events. The experiment shows that this method does not suffer from source dependence, i.e., the preference for betting on some events based on knowledge about the mechanism that generates them. However, it also highlights how additivity might be violated. This paper deduces the practical implications of these results.