Enhanced Credit Risk Management in Financial Institutions Using Regression Analysis

Credit risk assessment is crucial for financial institutions, including savings and credit cooperative societies, to evaluate borrower creditworthiness and mitigate default risks. Traditional predictive analytics methods, such as linear regression and decision trees have been widely used. However, they often fall short of capturing complex, non-linear relationships inherent in financial data, leading to suboptimal risk predictions. This study introduces an enhanced predictive analytics model that uses polynomial logistic regression, augmented with Recursive Feature Elimination (RFE) and ridge regression. The model captures the intricate dynamics between key risk factors such as interest rates, income stability, and collateral value to give better performance. The model was trained and validated on credit risk dataset from Kaggle which achieved an Area Under the Curve (AUC) of 0.95, indicating a strong capability to distinguish between defaulters and non-defaulters. Comparative analyses with alternative machine learning models that include XGBoost and Random Forest demonstrated that while these models offer high predictive accuracy, they often require extensive hyperparameter tuning and lack interpretability. In contrast, the proposed logistic regression model balanced predictive performance with interpretability and computational efficiency, making it a better for credit risk management in financial institutions. Supplementary Material File (refined formated paper 16.05.25 fin.docx) - Download - 198.04 KB Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License.

Authors Metrics & Citations Metrics Article Usage 336views 158downloads Citations Download citation Jemimah Kilonzo, Michael Kimwele, Erick Omuya. Enhanced Credit Risk Management in Financial Institutions Using Regression Analysis. Authorea. 08 May 2025. DOI: https://doi.org/10.22541/au.174673444.41758234/v1 DOI: https://doi.org/10.22541/au.174673444.41758234/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.