Vol: 15, Issue: 4 2025

Page: 40-55

Loan Default Prediction in Microfinance: A Comparative Analysis of Logistic Regression and Decision Forest at Grameen Bank

Risshit Sikka

Received Date: 2025-07-04

Accepted Date: 2025-07-14

Published Date: 2025-10-17

http://doi.org/10.37648/ijtbm.v15i04.003

Loan Defaulting has become a prevalent issue in financial systems worldwide, as it threatens institutional stability and increases credit risks. When borrowers fail to meet repayment obligations, financial institutions face increasing risk. In Bangladesh the problem of loan defaulting has become especially severe with default rates crossing over 20.20% and even higher in some commercial owned banks at 45.79% [1]. The aim of this study was to understand what differentiates microfinance institutes like Grameen Bank that boast a relatively low default rate at 4.32% as of July 2025 compared to the majority of Banks in Bangladesh [2]. To evaluate the characteristics of loan defaulters, this study analyses the predictive performance of two-class logistic regression and two-class decision forest models in forecasting common loan default characteristics among micro-borrowers of Grameen Bank in Bangladesh. This study contributes by applying a primary survey of 106 Grameen Borrowers . The key findings of this study reveals that Monthly Household Income, Total Number of Microloans, Age and People in Household are the key characteristics of loan defaulters. Additionally, the decision tree model overall outperformed the logistic regression model with a higher F1 score and less error.

References

• Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
• GeeksforGeeks. (2017, May 9). Logistic regression in machine learning. https://www.geeksforgeeks.org/machine-learning/understanding-logistic-regression/
• Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression. John Wiley & Sons, Inc.
• Kagan, J. (2024, July 31). Microfinance definition: Benefits, history, and how it works. Investopedia. https://www.investopedia.com/terms/m/microfinance.asp
• Khandker, S., Khalily, B., & Khan, Z. (1994). Is Grameen Bank sustainable? (Human Resources Development and Operations Policy Working Papers). The World Bank. https://documents1.worldbank.org/curated/en/658601468768006874/pdf/multi-page.pdf
• Khandani, A. E., Kim, A. J., & Lo, A. W. (2010). Consumer credit-risk models via machine-learning algorithms. Journal of Banking & Finance, 34(11), 2767–2787. https://doi.org/10.1016/j.jbankfin.2010.06.001
• Noman, A. H. M., Pervin, S., Chowdhury, M. M., & Hossain, M. A. (2015). Comparative performance analysis between conventional and Islamic banks in Bangladesh- an application of binary logistic regression. Asian Social Science, 11(21), 248–259. https://doi.org/10.5539/ass.v11n21p248
• Shin, T. (2023, November 7). Understanding feature importance in machine learning. Built In. https://builtin.com/data-science/feature-importance
• Staff Correspondent. (2025, June 15). Defaulted loan rises to Tk 4.2 trillion. Prothomalo. https://en.prothomalo.com/business/local/6jwomf618g
• Wang, Y., Wang, T., & Liu, J. (2020). A comparative assessment of credit risk model based on machine learning ——a case study of bank loan data. Procedia Computer Science, 174, 141–149. https://doi.org/10.1016/j.procs.2020.06.069