Statistical Inference as a Bargaining Game
Electronic Access:
Free Download. Use the free Adobe Acrobat Reader to view this PDF file
Summary:
This paper extends the analogy, previously established by Learner (1978a), between a Bayesian inference problem and an economics allocation problem to show that posterior modes can be interpreted as optimal outcomes of a bargaining game. This bargaining game, over a parameter value, is played between two players: the researcher (with preferences represented by the prior) and the data (with preferences represented by the likelihood).
Series:
Working Paper No. 2002/081
Subject:
English
Publication Date:
May 20, 2002
ISBN/ISSN:
9781451850376/1018-5941
Stock No:
WPIEA0812002
Pages:
13
Please address any questions about this title to publications@imf.org