Nowcasting Global Poverty
This paper evaluates different methods for nowcasting country-level poverty rates, including methods that apply statistical learning to large-scale country-level data obtained from the World Development Indicators and Google Earth Engine. The metho...
| Main Authors: | , , |
|---|---|
| Language: | English |
| Published: |
World Bank, Washington, DC
2021
|
| Subjects: | |
| Online Access: | http://documents.worldbank.org/curated/undefined/143231637760743360/Nowcasting-Global-Poverty http://hdl.handle.net/10986/36636 |
| Summary: | This paper evaluates different
methods for nowcasting country-level poverty rates,
including methods that apply statistical learning to
large-scale country-level data obtained from the World
Development Indicators and Google Earth Engine. The methods
are evaluated by withholding measured poverty rates and
determining how accurately the methods predict the held-out
data. A simple approach that scales the last observed
welfare distribution by a fraction of real GDP per capita
growth—a method that departs slightly from current World
Bank practice—performs nearly as well as models using
statistical learning on 1,000+ variables. This GDP-based
approach outperforms all models that predict poverty rates
directly, even when the last survey is up to five years old.
The results indicate that in this context, the additional
complexity introduced by applying statistical learning
techniques to a large set of variables yields only marginal
improvements in accuracy. |
|---|