Home / Current Issue / Paper 1719877
Credit-Risk and Portfolio Analytics for Growth-Oriented Businesses: Integrating Three-Statement Models, Delinquency Metrics, Scenario Analysis, and Executive Dashboards
Subject area: Science,Engineering and Technology · Area of research: Data Science & Finance
DOI: https://doi.org/10.64388/IREV10I1-1719877
Abstract
Growth-oriented businesses require external finance to scale working capital, expand operations, invest in fixed assets and absorb temporary revenue volatility. Yet lenders, portfolio managers and financial analysts often face an information problem: traditional credit files describe past repayment behaviour, while management accounts and three-statement models describe future operating capacity. This article develops an integrated Credit-Risk and Portfolio Analytics Framework for Growth-Oriented Businesses (CRPAF-GOB) that connects three-statement financial models, delinquency metrics, portfolio segmentation, probability of default (PD), loss given default (LGD), exposure at default (EAD), scenario analysis and executive dashboards. The framework is grounded in the literature on credit scoring, corporate failure prediction, SME finance, data-driven financial management, internal controls and model-risk governance. A public credit-scoring benchmark, Kaggle's Give Me Some Credit dataset, is selected as the external reference dataset because it includes borrower distress, delinquency, revolving-utilization, debt-ratio and income variables. Because that dataset does not contain full SME financial statements, the empirical demonstration extends the benchmark logic into a non-proprietary SME portfolio dataset of 9,000 observations with revenue, margin, liquidity, leverage, working-capital, DSCR, delinquency and exposure fields. The analysis shows that delinquency stage, DSCR, cash-buffer days, current ratio, debt ratio, revolving utilization, revenue growth and accounts-receivable days are material risk drivers. In the demonstration portfolio, the random-forest classifier achieved an AUC of 0.772, while logistic regression achieved an AUC of 0.753. Heat-map analysis identifies Hospitality, Retail and Early-growth accounts as higher expected-loss areas, while scenario testing shows that a combined 20 percent revenue shock and six percentage-point margin compression can increase expected loss materially. The article contributes a practical analytics architecture for lenders, finance operations teams and growth-oriented businesses that need disciplined credit selection, portfolio monitoring, early-warning controls and executive decision support.
Keywords
Credit Risk, SME Finance, Portfolio Analytics, Three-Statement Modeling, Delinquency Metrics, Probability Of Default, Loss Given Default, Scenario Analysis, Executive Dashboards, Growth-Oriented Businesses.
How to cite this paper
@article{1719877,
author = {James Sydney, Lucy Ganyani, Mabasa Masunungure, Munashe Naphtali Mupa},
title = {Credit-Risk and Portfolio Analytics for Growth-Oriented Businesses: Integrating Three-Statement Models, Delinquency Metrics, Scenario Analysis, and Executive Dashboards},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {1},
pages = {2183-2197},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1719877.pdf},
abstract = {Growth-oriented businesses require external finance to scale working capital, expand operations, invest in fixed assets and absorb temporary revenue volatility. Yet lenders, portfolio managers and financial analysts often face an information problem: traditional credit files describe past repayment behaviour, while management accounts and three-statement models describe future operating capacity. This article develops an integrated Credit-Risk and Portfolio Analytics Framework for Growth-Oriented Businesses (CRPAF-GOB) that connects three-statement financial models, delinquency metrics, portfolio segmentation, probability of default (PD), loss given default (LGD), exposure at default (EAD), scenario analysis and executive dashboards. The framework is grounded in the literature on credit scoring, corporate failure prediction, SME finance, data-driven financial management, internal controls and model-risk governance. A public credit-scoring benchmark, Kaggle's Give Me Some Credit dataset, is selected as the external reference dataset because it includes borrower distress, delinquency, revolving-utilization, debt-ratio and income variables. Because that dataset does not contain full SME financial statements, the empirical demonstration extends the benchmark logic into a non-proprietary SME portfolio dataset of 9,000 observations with revenue, margin, liquidity, leverage, working-capital, DSCR, delinquency and exposure fields. The analysis shows that delinquency stage, DSCR, cash-buffer days, current ratio, debt ratio, revolving utilization, revenue growth and accounts-receivable days are material risk drivers. In the demonstration portfolio, the random-forest classifier achieved an AUC of 0.772, while logistic regression achieved an AUC of 0.753. Heat-map analysis identifies Hospitality, Retail and Early-growth accounts as higher expected-loss areas, while scenario testing shows that a combined 20 percent revenue shock and six percentage-point margin compression can increase expected loss materially. The article contributes a practical analytics architecture for lenders, finance operations teams and growth-oriented businesses that need disciplined credit selection, portfolio monitoring, early-warning controls and executive decision support.},
keywords = {Credit Risk, SME Finance, Portfolio Analytics, Three-Statement Modeling, Delinquency Metrics, Probability Of Default, Loss Given Default, Scenario Analysis, Executive Dashboards, Growth-Oriented Businesses.},
month = {July},
doi = {https://doi.org/10.64388/IREV10I1-1719877}
}