The Predictability of Macroeconomic Fluctuations: Methods and Empirical Analysis

宏观经济波动的可预测性：方法与实证分析

• 批准号: 8910601
• 负责人: Mark Watson
• 金额: $ 17.22万
• 依托单位: National Bureau of Economic Research Inc
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1989
• 资助国家: 美国
• 起止时间: 1989-08-01 至 1992-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8910601&HistoricalAwards=false
• 关键词: Predictability; Macroeconomic; Fluctuations; Methods; Empirical

Much of what is known about the evolution of the U.S. economy is based on statistical descriptions of the behavior of individual aggregate time series or of the relations among key series. This project continues and expands previous work on quantifying the sources of aggregate fluctuations. It consists of three parts. The first is primarily empirical, and focuses on the role of human capital accumulation (learning-by-doing) as a mechanism whereby short-run fluctuations have permanent effects. The second is also primarily empirical and concerns the predictability of recessions and expansions (as opposed simply to fluctuations in levels of growth rates) and the implications of this predictability for economic theory. The third is primarily one of theoretical econometrics and develops techniques for inference about orders of integration of time series variables. This project makes important methodological and substantive contributions to economics. The first part of the project provides one of the first systematic empirical investigations of the link between long-run growth and shocks traditionally viewed as purely temporary such as transitory shifts in monetary policy or in preferences. This has important implications for our understanding of the sources and the transmission of economic fluctuations. The second part builds on recent joint work aimed at developing a revised Index of Leading Economic Indicators. It could provide better predictions of future recessions and expansions. The third part yields new tools for applying time series methods to economic problems.

关于美国经济演变的许多已知知识都是基于对单个总体时间序列的行为或关键序列之间关系的统计描述。该项目继续并扩大了先前关于量化总波动来源的工作。它由三个部分组成。第一个主要是经验性的，侧重于人力资本积累(边做边学)作为一种短期波动具有永久性影响的机制的作用。第二个也主要是经验性的，涉及衰退和扩张的可预测性(与增长率水平的波动相对)，以及这种可预测性对经济理论的影响。第三种主要是理论计量经济学，并开发了关于时间序列变量的积分阶数的推断技术。该项目对经济学做出了重要的方法论和实质性贡献。该项目的第一部分提供了对长期增长与传统上被视为纯粹暂时的冲击(如货币政策或偏好的暂时性转变)之间联系的首批系统性实证调查之一。这对于我们理解经济波动的来源和传导具有重要意义。第二部分建立在最近旨在制定经修订的领先经济指标指数的联合工作的基础上。它可以对未来的衰退和扩张提供更好的预测。第三部分提出了将时间序列方法应用于经济问题的新工具。