Planned intervention: On Wednesday April 3rd 05:30 UTC Zenodo will be unavailable for up to 2-10 minutes to perform a storage cluster upgrade.
Published August 21, 2021 | Version v1
Conference paper Open

New Methods and Novel Data for Estimating Seasonal Hunger

  • 1. University of Washington

Description

We develop a model to estimate hunger as a function of climate indicators and nightlight intensity, measures of which are publicly available and updated monthly. We then compare this model to estimates of hunger based solely on time of year and past hunger rates at that time of year, and a model based on crop yields from the LSMS-ISA survey. These comparisons demonstrate how a model based on monthly, publicly available data and trained on past hunger data compares to a model based on no data other than the past hunger data, and a model based on agricultural survey data. The model based on monthly, publicly available data performs almost as well as the model based on survey data, illustrating how new frequently available data sources can aid development stakeholders in combatting food insecurity.

Files

Files (46.2 kB)

Name Size Download all
md5:820e99f25b4b0f04b78dbd428714c126
46.2 kB Download