CAREER: Inference on Macroeconomic Heterogeneity

最近关于总量经济的研究试图理解总量经济绩效如何对单个经济单位产生不同的影响。这部分是由于更容易获得详细的微观数据集，计算能力增强且更便宜，部分是由于政策制定者对不平等的担忧日益加剧。虽然在理解不平等与总体经济活动之间联系的理论和实证方法方面取得了很大进展，但学术研究人员、中央银行和其他政策机构目前使用的统计方法并没有考虑到这些差异。该提案包括四个项目，这些项目将开发新的和现代的计量经济学工具，以改进对宏观经济结果差异效应的实证研究。该项目还将制作视频材料，向学生和从业者传授这些计量经济学工具。这项研究项目的结果将改善宏观经济政策，促进经济增长，减少不平等，从而改善普通美国人的福祉。这个CAREER研究计划将使用三个项目来开发新的计量经济学方法来研究宏观经济学中的异质性。第一个项目开发了一个具有正式覆盖范围的推理程序，以保证可视化多维横断面异质性，例如跨家庭或公司群体的动态响应概况的异质性。第二个项目提供了估计时间序列数据的时间异质性和非线性的流行方法的因果重新解释，例如脉冲响应的状态或符号依赖。第三个项目对脉冲响应估计器进行了大规模的模拟研究，并就如何在许多可用的程序之间进行选择提供了定量建议。最后，该项目提出了一项计划，编制关于现代宏观计量经济学方法的互动式教材集。除了对计量经济学理论的贡献外，本研究项目的结果还将改善宏观经济政策，促进经济增长，减少不平等，从而提高普通美国人的福祉。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Recent research on the aggregate economy tries to understand how aggregate economic performance has differential impacts on individual economic units. This is partly due to increased access to detailed micro-level data sets, increased and cheaper computing power, and partly due to growing concerns among policymakers about inequality. While much progress has been made in both theory and empirical methods of understanding the links between inequality and aggregate economic activity, current statistical methods used by academic researchers, central banks, and other policy institutions are not designed to account for these differences. This proposal consists of four projects that will develop new and modern econometric tools to improve empirical research on differential effects of macroeconomic outcomes. The project will also develop video materials to educate students and practitioners in these econometric tools. The results of this research project will improve macroeconomics policies that increase economic growth as well as reduce inequality, thus improve the well-being of the average American. This CAREER research proposal will use three projects to develop new econometric methods for studying heterogeneity in macroeconomics. The first project develops an inference procedure with formal coverage that guarantees visualizing multi-dimensional cross-sectional heterogeneity, such as heterogeneity in dynamic response profiles across groups of households or firms. The second project provides a causal reinterpretation of popular methods for estimating temporal heterogeneity and nonlinearities in time series data, such as state- or sign-dependence of impulse responses. The third project runs a large-scale simulation study of impulse response estimators and provides quantitative recommendations on how to choose between the many available procedures. Finally, the project proposes a plan for developing a collection of interactive educational materials on modern macro-econometric methods. Besides the contribution to econometric theory, the results of this research project will also improve macroeconomics policies and increase economic growth as well as reduce inequality, thus improve the well-being of the average American.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.