The Power of Financial Reporting Quality and Audit Quality in Optimizing Fraud Detection Models
Abbas Kolivand
1
(
Department of Accounting, Faculty of Management, North Tehran Branch, Islamic Azad University, Tehran, Iran
)
Mohammad Hassani
2
(
Assistant Professor, Department of Accounting & Auditing, Faculty of Management, North Tehran Branch, Islamic Azad University, Tehran, Iran
)
Mehran Matinfard
3
(
Assistant Professor, Department of Accounting, Faculty of Management, North Tehran Branch, Islamic Azad University, Tehran, Iran
)
Keywords: Fraud Detection, Logit Approach, Multiple Discriminate Analysis Approach,
Abstract :
The purpose of this research is to compare the logit and multiple discriminate analysis models of fraud detection by combining and focusing on the components of financial reporting and audit quality. In this study, profit manipulation with 5 indicators has been used to detect fraud and determine fraudulent and non-fraudulent companies: a) cumulative accruals, b) discretionary accruals, c) smoothing operating profit, d) net profit and e) gross profit . Eckel index (1988) was used to measure profit smoothing levels. The statistical population of the study consists of all companies listed in the Tehran Stock Exchange. In this study, the four factors of financial reporting quality are audit quality, fraud triangle and finally the role of fraud whistleblowers based on corporate governance index used to predict the likelihood of fraud, resulting to develop a fraud detection models based on logit and multiple discriminate analysis approach. Findings from a survey of 104 companies in the period 2010 to 2020, indicate that: a) Using variables such as audit quality, disclosure quality, fraud triangle and internal auditing can lead to increase the accuracy of the initial model of Benish, and b) The accuracy of the logit model is higher than the multiple discriminate analysis in cumulative accruals, discretionary accruals, smoothing operating profit and gross profit portfolios and it works more efficiently in distinguishing fraudulent and non-fraudulent companies.
