Article: Predictive power of ensemble learning in audit qualifications: insights from Indian firms Journal: International Journal of Process Management and Benchmarking (IJPMB) 2025 Vol.21 No.4 pp.499 - 537 Abstract: This study evaluates the effectiveness of ensemble learning methods – bagging, boosting, and stacking – in predicting audit qualifications among Indian firms. Traditional models often fall short because of the complexity of financial reporting data. To address these limitations, our research introduces a novel stacked model incorporating a meta-classifier and utilises a comprehensive dataset enriched with essential financial and non-financial metrics. We systematically compared the performance of seven different ensemble models with conventional classifiers. Our findings demonstrate that ensemble methods, especially our proposed stacking approach, significantly outperform standard models in key metrics such as recall, precision, and area under the curve (AUC). This study highlights the critical role of meticulous feature selection and model calibration in optimising predictive analytics. This research provides valuable insights for auditors, regulatory bodies, and policymakers, enhancing audit quality predictions and supporting financial reporting integrity in the Indian context. Inderscience Publishers - linking academia, business and industry through research

Title: Predictive power of ensemble learning in audit qualifications: insights from Indian firms

Authors: T. Shahana; Vilvanathan Lavanya; Aamir Rashid Bhat

Addresses: Department of Management Studies, National Institute of Technology Tiruchirappalli, Thuvakkudi, Trichy, 620015, India ' Department of Management Studies, National Institute of Technology Tiruchirappalli, Thuvakkudi, Trichy, 620015, India ' Department of Corporate Secretaryship Accounting and Finance, SRM Institute of Science and Technology, Kattankulathur, Chennai, 603203, India

Abstract: This study evaluates the effectiveness of ensemble learning methods – bagging, boosting, and stacking – in predicting audit qualifications among Indian firms. Traditional models often fall short because of the complexity of financial reporting data. To address these limitations, our research introduces a novel stacked model incorporating a meta-classifier and utilises a comprehensive dataset enriched with essential financial and non-financial metrics. We systematically compared the performance of seven different ensemble models with conventional classifiers. Our findings demonstrate that ensemble methods, especially our proposed stacking approach, significantly outperform standard models in key metrics such as recall, precision, and area under the curve (AUC). This study highlights the critical role of meticulous feature selection and model calibration in optimising predictive analytics. This research provides valuable insights for auditors, regulatory bodies, and policymakers, enhancing audit quality predictions and supporting financial reporting integrity in the Indian context.

Keywords: ensemble learning; audit qualifications; financial reporting; machine learning models; predictive analytics; Indian firms.

DOI: 10.1504/IJPMB.2025.149981

International Journal of Process Management and Benchmarking, 2025 Vol.21 No.4, pp.499 - 537

Received: 18 May 2024
Accepted: 31 May 2024
Published online: 20 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Predictive power of ensemble learning in audit qualifications: insights from Indian firms

Keep up-to-date