Mutual observability and the convergence of actions in a multi-person two-armed bandit model

成果类型:
Article
署名作者:
Aoyagi, M
署名单位:
Pennsylvania Commonwealth System of Higher Education (PCSHE); University of Pittsburgh
刊物名称:
JOURNAL OF ECONOMIC THEORY
ISSN/ISSBN:
0022-0531
DOI:
10.1006/jeth.1995.2450
发表日期:
1998
页码:
405-424
关键词:
two-armed bandit experimentation mutual observability Herd behavior
摘要:
This paper studies a model of a two-armed bandit played in parallel by two or more players. Players observe the actions of all other players, but not the outcome of their experiments. It is shown that if the parameters of the two arms (i.e., their success probabilities) are different by a fixed margin, all players eventually settle on the same arm with probability one in any Nash equilibrium of the game. (C) 1998 Academic Press.