Using Machine Learning to Identify Factors Contributing to Firms’ Bankruptcy: A Case Study of the Taiwanese Market

Chaudhary, Saumya (2023) Using Machine Learning to Identify Factors Contributing to Firms’ Bankruptcy: A Case Study of the Taiwanese Market. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Making the right financing decisions requires effective bankruptcy prediction on the part of financial institutions. In general, the two most significant aspects determining the prediction performance are the input variables (or features), such as financial ratios, and prediction methodologies, such as statistical and machine learning approaches. Even though numerous relevant publications have suggested innovative prediction methods, only a few have examined the crucial financial ratios that influence bankruptcy prediction. This research examines various statistical and machine-learning approaches to identify the important financial factors that can contribute to a company’s bankruptcy. Logistic Regression, Decision Tree, Random Forest, Naive Bayes, Support Vector, Balanced Random Forest, and Easy Ensemble classifiers are trained to evaluate their efficacy on the real-world Taiwanese bankruptcy dataset. Sensitivity, specificity, type 1 and type 2 error rates, and receiver operating characteristics values are the metrics used to assess the models’ predictability. Among the ensemble methods, Balanced Random Forest and Easy Ensemble outperformed the others, and eleven financial ratios were deemed important: ROA (C) and depreciation before interest, Degree of Financial Leverage, Borrowing dependency, Debt ratio, Non-industry income and expenditure/revenue, Equity to Liability ratio, Interest Coverage Ratio, Total income/expense ratio, Interest Expense Ratio, Net Value Per Share, and Total debt/Total net worth ratio.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Byrne, Brian UNSPECIFIED
Subjects:	H Social Sciences > HG Finance > Credit. Debt. Loans. > Bankruptcy H Social Sciences > HG Finance > Fintech T Technology > T Technology (General) > Information Technology > Fintech Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in FinTech
Depositing User:	Tamara Malone
Date Deposited:	02 Aug 2024 10:42
Last Modified:	02 Aug 2024 10:42
URI:	https://norma.ncirl.ie/id/eprint/7015

Actions (login required)

View Item