Nonparametric adaptive learning with feedback
成果类型：
Article
署名作者：
Chen, XH; White, H
署名单位：
University of Chicago; University of California System; University of California San Diego
刊物名称：
JOURNAL OF ECONOMIC THEORY
ISSN/ISSBN：
0022-0531
DOI：
10.1006/jeth.1998.2432
发表日期：
1998
页码：
190-222
关键词：
摘要：
Recently macroeconomists and game theorists have dropped the mutual consistency assumption for rational expectations equilibrium (REE) or Nash equilibrium (NE), considering instead plausible adaptive agent learning behaviors yielding sensible REE or NE. But when agents do not have detailed knowledge of the relevant equilibrium relations, they can easily arrive at incorrect belief equilibria which are not rational and have no optimality properties [14]. REE and NE thus lose plausibility as outcomes of learning. Here we study nonparametric adaptive learning methods that enable agents to eventually learn the relevant equilibrium relations, leading to sensible REE and NE. (C) 1998 Academic Press.