Is Random Forest a Superior Methodology for Predicting Poverty? An Empirical Assessment

Sohnesen, Thomas Pave; Stender, Niels

doi:10.1002/pop4.169

Cited by 32 publications

(22 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper provides a comprehensive analysis on the drivers of poverty via the application of various estimation methods, such as probit, tobit, and machine learning algorithms. While it is widely recognised in the poverty literature that traditional methods such as probit or tobit can contribute to understanding the main factors influencing poverty status, the application of machine learning techniques in this context is limited to only a few countries (Sohnesen and Stender, 2016;Zhao et al, 2019). Because machine learning, unlike probit or tobit, offers a nonparametric and flexible way to model poverty status, we hope to develop a superior povertyidentification technique that not only results in higher predictive power but also reveals more clearly the underlying empirical relationships considered.…”

Section: Motivation Behind Empirical Methodsmentioning

confidence: 99%

Winning the War? New Evidence on the Measurement and the Determinants of Poverty in the United States

Weber¹,

Bergant²,

Medici³

2022

IMF Working Papers

View full text Add to dashboard Cite

and seminar participants at the IMF for their helpful comments. We are grateful to Bruce Meyer and James Sullivan for generously sharing their codes and helpful discussions on the construction of their poverty measures. All remaining errors are our own. The views expressed in IMF Working Papers are those of the authors and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.

show abstract

Section: Motivation Behind Empirical Methodsmentioning

confidence: 99%

Winning the War? New Evidence on the Measurement and the Determinants of Poverty in the United States

Weber¹,

Bergant²,

Medici³

2022

IMF Working Papers

View full text Add to dashboard Cite

show abstract

“…While the selection of variables based on LASSO, present better results compared to RF in terms of reducing the inclusion error (non-poor households classified as poor). Sohnesen & Stender (2017) also, use LASSO and RF to predict poverty using data of a period of one year. The results indicate that RF is a good predictor of poverty and obtains more robust estimates compared to Linear Regression estimators.…”

Section: Related Work Social Science and Machine Learningmentioning

confidence: 99%

Machine learning approach for multidimensional poverty estimation

Guaraca

Castro

Pallaroso

et al. 2021

rte

View full text Add to dashboard Cite

In the social sciences, a theoretical analysis has predominated in its research. The scarcity of data and its difficulty in collecting and storing it, has been the main limitation for the social sciences to adopt quantitative approaches. However, the large amount of information generated in recent years, mainly through the use of the Internet, has allowed the social sciences to include more and more quantitative analysis. This study proposes the use of technologies such as Machine Learning (ML) are the answers to solving this data scarcity. The objective is to estimate the multidimensional poverty index at the personal level in a particular territory of Ecuador by using Machine Learning (ML) regression models based on a limited amount of data for training. Ten ML models are compared, such as linear, regularized, and assembled models and Random Forest performs outstandingly against the other models. An error of 7.5% was obtained in the cross-validation and 7.48% with the test data set. The estimates are compared with statistical approximations of the MPI in a geographical area and it is obtained that the average MPI estimated by the model compared to the average reported by the statistical studies differs by 1%.

show abstract

“…McBride and Nichols (2016) seminal application of machine learning to poverty targeting is the first to use an approach that is explicitly designed to improve out-of-sample prediction. However, along with Sohnesen and Stender (2017), they employ a black-box random forest methodology. 29 Mark Schreiner's insightful and widely-used approach to designing proxy-means tests (e.g.…”

Section: Pragmatic Drought-contingent Targeting Strategymentioning

confidence: 99%

Adaptive Safety Nets for Rural Africa: Drought-Sensitive Targeting with Sparse Data

Báez¹,

Kshirsagar²,

Skoufias³

2019

View full text Add to dashboard Cite

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.

show abstract

Is Random Forest a Superior Methodology for Predicting Poverty? An Empirical Assessment

Cited by 32 publications

References 17 publications

Winning the War? New Evidence on the Measurement and the Determinants of Poverty in the United States

Winning the War? New Evidence on the Measurement and the Determinants of Poverty in the United States

Machine learning approach for multidimensional poverty estimation

Adaptive Safety Nets for Rural Africa: Drought-Sensitive Targeting with Sparse Data

Contact Info

Product

Resources

About