Forecasting GDP Growth Rates Using Accounting Earnings: A Large Panel Microdata Approach

Economists and econometricians typically use aggregate economic and financial variables for gross domestic product (GDP) prediction. However, aggregation often results in a loss of valuable information, diminishing key features such as heterogeneity, interactions, nonlinearity, and structural breaks. We propose a novel microforecasting approach, using large panel data of firm accounting earnings from corporate financial reports to forecast GDP. By employing machine learning methods, we can effectively exploit this large microlevel information set to achieve substantially more accurate GDP forecasts. Our findings highlight the advantages and potential of utilizing microlevel data for macroprediction, diverging from the conventional macroforecasting paradigm that relies on aggregate data to forecast macrovariables. This paper was accepted by Will Cong for the Special Issue on Digital Finance. Funding: Y. Hong is supported by the MOE Social Sciences Innovative Group on Complex Systems Modeling in Economic Management in the Era of Digital Intelligence, University of Chinese Academy of Sciences [Grant E5820801] and the National Science Foundation of China (NSFC) Basic Science Center Project “Econometric Modeling and Economic Policy Studies” [Grant 71988101]. N. Huang is supported by the NSFC [Grant 72473166]. Y. Wang is supported by the Shenzhen Municipal Government, National Natural Science Foundation of China [Grant 20220810114654001]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2025.01549 .