Unveiling the Impact of Key Determinants on Microloan Allocation Using Machine Learning

Abstract

Microfinance initiatives have emerged as crucial tools in poverty alleviation, particularly in developing countries.  These initiatives extend small loans to underserved populations and marginalized communities, as a means to improve their financial standing.  This research leverages machine learning models to analyze a large microfinance initiative and find the key variables that drive micro-loan allocation. By conducting an in-depth feature analysis of a comprehensive microloan dataset, we aimed to pinpoint key variables that impact the allocation and size of microloans. Our study finds that the repayment period, lender networks, and the geographic location of borrowers significantly influence the loan allocation and size. The insights from our study can provide strategic guidance to microfinance programs in their loan allocation decisions.  More specifically, lending decisions can be targeted more effectively to maximize impact of microfinance programs. In addition, this research also demonstrates the application of machine learning models in the realm of microfinance.