Audit Risk Prediction and Fraudulent Firm Classification Using Business Analytics

Abstract

Audit risk prediction has become increasingly important as auditors face growing volumes of complex financial and operational data. Business analytics offers a structured approach for identifying high-risk firms and supporting fraudulent firm classification through measurable audit-risk indicators. This study investigates the role of audit-risk indicators in distinguishing risky firms from non-risky firms. A quantitative research design was adopted, using descriptive statistics, comparative analysis, correlation analysis, and classification-based assessment. The analysis considered major audit-risk dimensions, including inherent risk, monetary exposure, total audit parameter exposure, component risk indicators, control risk, and overall audit risk. The results reveal that risky firms recorded substantially higher values for inherent risk, monetary exposure, total audit parameter exposure, and overall audit risk compared with non-risky firms. Correlation findings indicate that composite audit score, monetary exposure score, audit parameter scores, control risk, district-level loss, and inherent risk were meaningfully associated with final risk classification. The classification assessment further confirmed that audit-risk indicators can effectively separate risky firms from non-risky firms. The findings highlight the practical value of business analytics in strengthening audit planning, improving risk-based screening, and supporting data-driven decision-making among auditors, managers, regulators, and governance stakeholders. The study contributes to accounting and auditing literature by demonstrating how structured audit-risk indicators can support fraudulent firm classification and audit risk assessment.