Using Neural Networks to Predict Micro-Spatial Economic Growth
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 1.2km and 2.4km (where the average US county has dimension of 55.6km), 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.
This project was funded through the support of the Russell Sage Foundation program on Computational Social Science. This funding did not involve any clearance. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
Arman Khachiyan & Anthony Thomas & Huye Zhou & Gordon Hanson & Alex Cloninger & Tajana Rosing & Amit K. Khandelwal, 2022. "Using Neural Networks to Predict Microspatial Economic Growth," American Economic Review: Insights, American Economic Association, vol. 4(4), pages 491-506, December. citation courtesy of