Asymptotic Learning with Ambiguous Information

• 成果类型: Article
• 署名作者: Reshidi, Pellumb; Thereze, Joao; Zhang, Mu
• 署名单位: State University System of Florida; Florida State University; Duke University; University of Michigan System; University of Michigan
• 刊物名称: AMERICAN ECONOMIC JOURNAL-MICROECONOMICS
• ISSN/ISSBN: 1945-7669
• DOI: 10.1257/mic.20230142
• 发表日期: 2025
• 页码: 244-288
• 关键词: convergence; DESIGN; models

摘要

We study asymptotic learning when the decision-maker faces ambiguity in the precision of her information sources. She aims to estimate a state and evaluates outcomes according to the worst-case scenario. Under prior-by-prior updating, we characterize the set of asymptotic posteriors the decision-maker entertains, which consists of a continuum of degenerate distributions over an interval. Moreover, her asymptotic estimate of the state is generically incorrect. We show that even a small amount of ambiguity may lead to large estimation errors and illustrate how an econometrician who learns from observing others' actions may over-or underreact to information. (JEL D82, D83)