Skip to content

Latest commit

 

History

History
5 lines (4 loc) · 613 Bytes

File metadata and controls

5 lines (4 loc) · 613 Bytes

Targeting-geospatial-data-Malawi.

Collaboration with David Newhouse, Senior Economist at World Bank.

This repository contains the code for pulling geospatial from Google Earth Engine, and combine it with gridded shapefiles of Malawi using R. The R code also includes XGBoost models to predict benchmark welfare and predicted welfare using census/survey informationcombined with geospatial data. It contains Stata do-files to process Malawi data, including 20% census extract, LSMS surey, UBR (admin data) and run Lasso models to predict welfare combining these traditional data sources with geospatial data.