Error-Correction Reinterpretation and Efficient Estimation of Dynamic Stochastic General Equilibrium Models

动态随机一般均衡模型的误差修正重新解释和有效估计

• 批准号: 1529151
• 负责人: Jean-Francois Richard
• 金额: $ 11.23万
• 依托单位: University of Pittsburgh
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-06-01 至 2017-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1529151&HistoricalAwards=false
• 关键词: Error; Correction; Reinterpretation; Efficient; Estimation

Dynamic Stochastic General Equilibrium (DSGE) models are workhorses of empirical macroeconomics, and are widely used for policy analysis. The PI's research seeks to reinterpret DSGE models within a more flexible mathematical framework. The goal is to provide methods that address some key deficiencies of DSGE models and provide a more robust framework for understanding business cycle fluctuations and the effects of government policies. The PI plans to develop a method that is in some ways similar to the Bank of England Quarterly Model, in which steady state core solutions are embedded within Error Correction Mechanism (ECM) equations using detrended data. There is a key difference; rather than detrending data and constructing ECM equations for core solutions, the PI plans to use as ECM targets balanced growth (log) ratios between core solutions. There are two reasons for proceeding in this way. First, these rations are obvious candidates for co-integrating relationships, and second, the advantages of formally accounting for (near) unit roots are well understood. They robustify inference of key structural parameters and play a key role in separating persistent from less persistent movements in the data. The PI will also continue to develop numerically efficient filtering techniques applicable to ECM/DSGE models.

动态随机一般均衡（DSGE）模型是实证宏观经济学的中坚力量，被广泛用于政策分析。PI的研究试图在更灵活的数学框架内重新解释DSGE模型。目标是提供方法，解决DSGE模型的一些关键缺陷，并为理解商业周期波动和政府政策的影响提供一个更有力的框架。PI计划开发一种在某些方面类似于英格兰银行季度模型的方法，其中稳态核心解决方案嵌入使用趋势数据的纠错机制（ECM）方程中。这里有一个关键的区别；PI计划使用核心解决方案之间的平衡增长（对数）比率作为ECM目标，而不是对数据进行趋势分析和构建核心解决方案的ECM方程。这样做有两个原因。首先，这些定量是协整关系的明显候选者，其次，正式计算（近）单位根的优势是很容易理解的。它们增强了关键结构参数的推断，并在分离数据中的持久运动和非持久运动方面发挥了关键作用。PI还将继续开发适用于ECM/DSGE模型的数值高效滤波技术。