Can Medium-Resolution Satellite Imagery Measure Economic Activity at Small Geographies? Evidence from Landsat in Vietnam
This study explores the potential and the limits of medium-resolution satellite data as a proxy for economic activity at small geographic units. Using a commune-level dataset from Vietnam, it compares the performance of commonly used nightlight dat...
| Main Authors: | , , |
|---|---|
| Format: | Working Paper |
| Language: | English |
| Published: |
World Bank, Washington, DC
2019
|
| Subjects: | |
| Online Access: | http://documents.worldbank.org/curated/en/168331576612750230/Can-Medium-Resolution-Satellite-Imagery-Measure-Economic-Activity-at-Small-Geographies-Evidence-from-Landsat-in-Vietnam http://hdl.handle.net/10986/33057 |
| Summary: | This study explores the potential and
the limits of medium-resolution satellite data as a proxy
for economic activity at small geographic units. Using a
commune-level dataset from Vietnam, it compares the
performance of commonly used nightlight data and higher
resolution Landsat imagery which measures daytime light
reflection. The analysis suggests that Landsat outperforms
nighttime lights at predicting enterprise counts,
employment, and expenditure in simple regression models. A
parsimonious combination of the first two moments of the
Landsat spectral bands can explain a reasonable share of the
variation in economic activity in the cross-section. There
is however poor prediction power of either satellite measure
for changes over time. |
|---|