Using Neural Networks to Predict Microspatial Economic Growth

Arman Khachiyan, Anthony Thomas, Huye Zhou, Gordon Hanson, Alex Cloninger, Tajana Rosing and Amit Khandelwal

American Economic Review: Insights, 2022, vol. 4, issue 4, 491-506

Abstract: We apply deep learning to daytime satellite imagery to predict changes in income and population at high spatial resolution in US data. For grid cells with lateral dimensions of 1.2 km and 2.4 km (where the average US county has dimension of 51.9 km), our model predictions achieve R2 values of 0.85 to 0.91 in levels, which far exceed the accuracy of existing models, and 0.32 to 0.46 in decadal changes, which have no counterpart in the literature and are 3–4 times larger than for commonly used nighttime lights. Our network has wide application for analyzing localized shocks.

JEL-codes: C45 R11 R23 (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
https://www.aeaweb.org/doi/10.1257/aeri.20210422 (application/pdf)
https://doi.org/10.3886/E158002V1 (text/html)
https://www.aeaweb.org/doi/10.1257/aeri.20210422.appx (application/pdf)
https://www.aeaweb.org/doi/10.1257/aeri.20210422.ds (application/zip)
Access to full text is restricted to AEA members and institutional subscribers.

Related works:
Working Paper: Using Neural Networks to Predict Micro-Spatial Economic Growth (2021)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:aea:aerins:v:4:y:2022:i:4:p:491-506

Ordering information: This journal article can be ordered from
https://www.aeaweb.org/journals/subscriptions

DOI: 10.1257/aeri.20210422

Access Statistics for this article

American Economic Review: Insights is currently edited by Amy Finkelstein

More articles in American Economic Review: Insights from American Economic Association Contact information at EDIRC.
Bibliographic data for series maintained by Michael P. Albert ().