Data and Code for: Using Neural Networks to Predict Micro-Spatial Economic Growth

Principal Investigator(s):  View help for Principal Investigator(s) Arman Khachiyan, UC San Diego; Anthony Thomas, UC San Diego; Huye Zhou, UC San Diego; Gordon Hanson, Harvard University; Alex Cloninger, UC San Diego; Tajana Rosing, UC San Diego; Amit Khandelwal, Columbia University

Version:  View help for Version V1

Name File Type Size Last Modified
  code 12/23/2021 09:57:PM
  data 04/13/2022 03:58:PM
  weights 04/09/2022 09:47:PM
README.md text/x-web-markdown 23.9 KB 04/26/2022 07:54:PM
environment.yml text/plain 259 bytes 04/11/2022 08:36:AM
setup.sh application/x-sh 147 bytes 04/11/2022 08:37:AM

Project Citation: 

Khachiyan, Arman, Thomas, Anthony, Zhou, Huye, Hanson, Gordon, Cloninger, Alex, Rosing, Tajana, and Khandelwal, Amit. Data and Code for: Using Neural Networks to Predict Micro-Spatial Economic Growth. Nashville, TN: American Economic Association [publisher], 2022. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2022-11-21. https://doi.org/10.3886/E158002V1

Project Description

Summary:  View help for Summary 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 1.2sq-km and 2.4sq-km (where the average US county has dimension of 55.6km), our model predictions achieve R-sq 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.
Funding Sources:  View help for Funding Sources Russell Sage Foundation (G-2196)

Scope of Project

Subject Terms:  View help for Subject Terms spatial economics; income measurement; remote sensing
JEL Classification:  View help for JEL Classification
      R10 General Regional Economics (includes Regional Data)
Geographic Coverage:  View help for Geographic Coverage Contiguous United States
Time Period(s):  View help for Time Period(s) 1/2000 – 12/2020
Collection Date(s):  View help for Collection Date(s) 1999 – 2020
Universe:  View help for Universe Analysis is conducted on square images of neighborhoods in the contiguous United States.
Data Type(s):  View help for Data Type(s) administrative records data; census/enumeration data; geographic information system (GIS) data; images: photographs, drawings, graphical representations
Collection Notes:  View help for Collection Notes Please see our README.md file and our manuscript for details on how each dataset was collected and applied in our study.

Methodology

Response Rate:  View help for Response Rate n/a
Sampling:  View help for Sampling All neighborhoods above a population density threshold were included in our analysis.
Data Source:  View help for Data Source Landsat 7 imagery via Google Earth Engine
Census Estimates via IPUMS NHGIS
Collection Mode(s):  View help for Collection Mode(s) face-to-face interview; mail questionnaire; remote sensing
Scales:  View help for Scales No scales were used.
Weights:  View help for Weights Weights are not used in our analysis.
Unit(s) of Observation:  View help for Unit(s) of Observation The unit of observation is a neighborhood (image) in a year.
Geographic Unit:  View help for Geographic Unit The geographic unit is a neighborhood (image).

Related Publications

Download this project

Published Versions

V1 [2022-11-21]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.