Addressing Imbalance in Poverty Classification: A SMOTE-Enabled Statistical Analysis Approach

Accurately assessing poverty is vital for policy development and growth planning. Using data from the NITI Aayog-India Multinational Poverty Index Progress Review 2023, this study assesses how sophisticated statistical techniques and data-balancing procedures handle difficulties in imbalanced datase...

Full description

Saved in:
Bibliographic Details
Main Authors: Shushant Hatwar, Yogalakshmi Thangaraj, Sujatha Vishnumoorthy
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:Journal of Mathematics
Online Access:http://dx.doi.org/10.1155/jom/5357997
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accurately assessing poverty is vital for policy development and growth planning. Using data from the NITI Aayog-India Multinational Poverty Index Progress Review 2023, this study assesses how sophisticated statistical techniques and data-balancing procedures handle difficulties in imbalanced datasets for poverty detection. For resolving imbalances, important techniques include the Huber regressor, Theil–Sen estimator, canonical correlation analysis (CCA), logistic regression, and SMOTE. While CCA identified important determinants of poverty, SMOTE significantly improved the accuracy of logistic regression. The Theil–Sen estimator fought off outliers, while the Huber regressor successfully handled extreme data. The results highlight the value of improved models for classifying poverty in order to facilitate focused initiatives to reduce it.
ISSN:2314-4785