òòò½Íø Papers and Proceedings
ISSN 2574-0768 (Print) | ISSN 2574-0776 (Online)
Nowcasting Distributional National Accounts for the United States: A Machine Learning Approach
òòò½Íø Papers and Proceedings
vol. 115,
May 2025
(pp. 79–84)
Abstract
Inequality statistics are usually calculated from high-quality microdata, typically available with a one-to-two-year lag. In turbulent times, policymakers and data users need timely estimates. We use an elastic net to nowcast the overall Gini coefficient and quintile-level income shares of personal income published by the US Bureau of Economic Analysis. The nowcasts (2020–2023) predict turning points with at least 90 percent accuracy across all metrics and significantly less error than naive models. We find that we can create advance inequality estimates approximately one month after the end of the calendar year, reducing the present lag by almost a year.Citation
Cornwall, Gary, and Marina Gindelsky. 2025. "Nowcasting Distributional National Accounts for the United States: A Machine Learning Approach." òòò½Íø Papers and Proceedings 115: 79–84. DOI: 10.1257/pandp.20251105Additional Materials
JEL Classification
- C45 Neural Networks and Related Topics
- C53 Forecasting Models; Simulation Methods
- E01 Measurement and Data on National Income and Product Accounts and Wealth; Environmental Accounts
- E23 Macroeconomics: Production
- E27 Macroeconomics: Consumption, Saving, Production, Employment, and Investment: Forecasting and Simulation: Models and Applications