Development of a Predictive Model in Mapping the Payment Trajectories of Low-income Borrowers Using Logistic Regression Analysis

Abstract

One of the main challenges faced by low-income borrowers is the high incidence of loan default and delinquency. Consequently, there is a need to identify the factors that influence loan repayment behavior and develop a predictive model that can forecast the likelihood of default or delinquency. To address this issue, this study developed a predictive model that accurately predicts loan repayment behavior, so that the potential negative impacts on both borrowers and lenders can be mitigated, promoting financial stability and inclusive growth. This study employed a quantitative research design, specifically utilizing logistic regression analysis, to develop a predictive model for mapping the payment trajectories of the 242 low-income borrowers in Occidental Mindoro, Philippines, who were selected using stratified random sampling. The findings revealed that the developed model effectively captures the variations in low-income borrowers' payment trajectories, accurately distinguishing between making timely payments or failing to make timely payments. By understanding the demographic characteristics, income sources, and factors that influence repayment behavior, tailored financial interventions can be designed to provide affordable credit options, promote financial education, and support borrowers in making timely payments. A specific recommendation is to develop comprehensive financial education programs tailored specifically for low-income borrowers. These programs should focus on improving financial literacy, budgeting skills, and informed decision-making by addressing the knowledge gap identified among this population.