A new GLO Discussion Paper concludes that the quality of predictions and the choice of the optimal prediction model are dependent on the distribution of observed and unobserved incomes, the poverty line, the choice of objective function and policy preferences, and various other modeling choices.
GLO Discussion Paper No. 1260, 2023
GLO Fellow Paolo Verme
Author Abstract: Poverty prediction models are used by economists to address missing data issues in a variety of contexts such as poverty profiling, targeting with proxy-means tests, cross-survey imputations such as poverty mapping, or vulnerability analyses. Based on the models used by this literature, this paper conducts an experiment by artificially corrupting data with different patterns and shares of missing incomes. It then compares the capacity of classic econometric and machine learning models to predict poverty under these different scenarios. It finds that the quality of predictions and the choice of the optimal prediction model are dependent on the distribution of observed and unobserved incomes, the poverty line, the choice of objective function and policy preferences, and various other modeling choices. Logistic and random forest models are found to be more robust than other models to variations in these features, but no model invariably outperforms all others. The paper concludes with some reflections on the use of these models for predicting poverty.
Featured image: Mika-Baumeister-on-Unsplash
Vol. 36, Issue 3, July 2023: Journal of Population Economics (JOPE) 26 articles.
JOPE has CiteScore 6.5 (2021, LINK) & Impact Factor 4.7 (2021, LINK)
GLO Discussion Papers are research and policy papers of the GLO Network which are widely circulated to encourage discussion. Provided in cooperation with EconStor, a service of the ZBW – Leibniz Information Centre for Economics, GLO Discussion Papers are among others listed in RePEc (see IDEAS, EconPapers). Complete list of all GLO DPs – downloadable for free.
The Global Labor Organization (GLO) is an independent, non-partisan and non-governmental organization that functions as an international network and virtual platform to stimulate global research, debate and collaboration.