Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Journal of Advanced Sciences and Mathematics Education
Sherena Wahyutari
Universitas Muhammadiyah Surakarta
Chemistry Education Mathematics Physics Social Sciences

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Financial signals and governance in fraud detection: Evidence from Indonesia’s energy sector using logistic regression and random forest Sherena Wahyutari; Triyono; Banu Witono
Journal of Advanced Sciences and Mathematics Education Vol. 5 No. 2 (2025): Journal of Advanced Sciences and Mathematics Education
Publisher : CV. FOUNDAE

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58524/jasme.v5i2.986

Abstract

Background: Fraud in financial reporting still appears in Indonesia’s energy industry, a field where complex operations often conceal early signs of misstatements. In many cases, day-to-day financial patterns reveal more dependable clues than the formal structure of corporate governance. Aim: The study examines how governance features and financial indicators contribute to identifying possible manipulation in financial statements and evaluates the predictive strength of logistic regression compared with Random Forest. Method: The analysis uses 171 firm-year observations from energy companies listed on the Indonesia Stock Exchange between 2022 and 2024. Potential irregularities were screened using the Beneish M-Score. Governance information covers the share of independent commissioners, CEO duality, board size, and board meeting frequency, while profitability, operating cash flow, and sales growth serve as the financial indicators. Both logistic regression and Random Forest were employed, and their performance was reviewed through accuracy, sensitivity, specificity, and AUC values. Results: Governance variables showed no meaningful link to the likelihood of fraud. In contrast, profitability, operating cash flow, and sales growth consistently appeared as significant indicators. Logistic regression produced stronger classification results, reaching 79.4 percent accuracy with an AUC of 0.814, compared with Random Forest’s 70.6 percent accuracy and 0.731 AUC. Conclusion: Financial indicators proved more reliable than governance characteristics in signaling possible fraudulent reporting. Logistic regression also offered steadier predictive behavior than Random Forest, making it particularly useful for monitoring firms in the Indonesian energy sector.
Co-Authors Banu Witono Triyono