GENERALIZED STOCHASTIC GRADIENT LEARNING

• 成果类型: Article
• 署名作者: Evans, George W.; Honkapohja, Seppo; Williams, Noah
• 署名单位: University of Wisconsin System; University of Wisconsin Madison; University of Oregon; University of St Andrews
• 刊物名称: INTERNATIONAL ECONOMIC REVIEW
• ISSN/ISSBN: 0020-6598
• DOI: 10.1111/j.1468-2354.2009.00578.x
• 发表日期: 2010
• 页码: 237-262
• 关键词: monetary-policy; nash inflation; STABILITY; expectations; CONVERGENCE; FRAMEWORK; beliefs; rules; MODEL

摘要

We study the properties of the generalized stochastic gradient (GSG) learning in forward-looking models. GSG algorithms are a natural and convenient way to model learning when agents allow for parameter drift or robustness to parameter uncertainty in their beliefs. The conditions for convergence of GSG learning to a rational expectations equilibrium are distinct from but related to the well-known stability conditions for least squares learning.