Promoting Financial Inclusion: Evaluating CTGAN-Based Synthetic Alternative Data for the Credit-Invisible

Odufuwa, Moyosoreoluwa Monsurat Ayomide (2025) Promoting Financial Inclusion: Evaluating CTGAN-Based Synthetic Alternative Data for the Credit-Invisible. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

The availability of high-quality credit-scoring data is essential for the development and validation of predictive models, yet privacy and regulatory constraints often limit access to such datasets. This study investigates the use of Conditional Tabular Generative Adversarial Networks (CTGAN) to generate synthetic credit-scoring data as a potential substitute for original datasets, particularly in research on alternative credit scoring for credit-invisible populations. The proposed approach involves minimal preprocessing of the original dataset to maintain fidelity, followed by CTGAN training to capture complex feature distributions and dependencies. The resulting synthetic dataset is evaluated for statistical similarity, predictive utility, and fairness. Results show high Pearson (0.8846) and Spearman (0.8335) correlation similarities, with tree-based classifiers trained on synthetic data achieving competitive ROC-AUC scores compared to those trained on real data. However, fairness metrics reveal that CTGAN reproduces much of the bias present in the original dataset. The findings suggest that while CTGAN-generated synthetic data holds strong promise as a privacy-preserving resource for academic research and early-stage prototyping, it cannot fully replace original datasets in production-grade applications without targeted efforts to address residual distributional gaps and embedded bias.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Yaqoob, Abid UNSPECIFIED
Subjects:	H Social Sciences > HG Finance Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HG Finance > Credit. Debt. Loans.
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	26 Aug 2026 10:14
Last Modified:	26 Aug 2026 10:14
URI:	https://norma.ncirl.ie/id/eprint/9654

Actions (login required)

View Item