<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>Considering the burning problem of corruption and non-transparency of public enterprises in the Federation of Bosnia and Herzegovina (FBiH), the paper aims to investigate whether the Beneish M-score model can be used to predict inaccurate financial statements. Where, the cause of inaccurate financial statements are intentional or unintentional errors.</p>
<p>On a sample of 200 financial statements of public enterprises and related audit reports issued by the Audit Office of the Institutions in FBiH, we made a link between the Beneish M score model with its partial indicators (DSRI, GMI, AQI, SGI, DEPI, SGAI, LVGI, TATA) and four types of opinions: positive, opinion with distraction, negative and refraining from giving opinions. The research was conducted using descriptive statistics and an artificial neural network with the “scaled conjugate gradient backpropagation (trainscg)” algorithm for pattern recognition and classification. The research results show that it is possible on the basis of 8 partial indicators (DSRI, GMI, AQI, SGI, DEPI, SGAI, LVGI, TATA) i.e. 24 balance sheet position for their calculation, predict the auditor’s opinion on the quality of financial statements of public companies with an accuracy ranging between 98 and 100% in repeated procedures. The results of the research have their practical usefulness and can serve to researchers, creditors, customers, suppliers and state auditors in planning resources and priorities for performing financial audits at public companies in the FBiH.</p>
</abstract>

Considering the burning problem of corruption and non-transparency of public enterprises in the Federation of Bosnia and Herzegovina (FBiH), the paper aims to investigate whether the Beneish M-score model can be used to predict inaccurate financial statements. Where, the cause of inaccurate financial statements are intentional or unintentional errors.
On a sample of 200 financial statements of public enterprises and related audit reports issued by the Audit Office of the Institutions in FBiH, we made a link between the Beneish M score model with its partial indicators (DSRI, GMI, AQI, SGI, DEPI, SGAI, LVGI, TATA) and four types of opinions: positive, opinion with distraction, negative and refraining from giving opinions. The research was conducted using descriptive statistics and an artificial neural network with the “scaled conjugate gradient backpropagation (trainscg)” algorithm for pattern recognition and classification. The research results show that it is possible on the basis of 8 partial indicators (DSRI, GMI, AQI, SGI, DEPI, SGAI, LVGI, TATA) i.e. 24 balance sheet position for their calculation, predict the auditor’s opinion on the quality of financial statements of public companies with an accuracy ranging between 98 and 100% in repeated procedures. The results of the research have their practical usefulness and can serve to researchers, creditors, customers, suppliers and state auditors in planning resources and priorities for performing financial audits at public companies in the FBiH.

Predicting Auditor’s Opinion on Financial Statements of Public Enterprises Based on Indicators of the Beneish M-score Model

Journal of Forensic Accounting Profession

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Considering the burning problem of corruption and non-transparency of public enterprises in the Federation of Bosnia and Herzegovina (FBiH), the paper aims to...

Predicting Auditor’s Opinion on Financial Statements of Public Enterprises Based on Indicators of the Beneish M-score Model

Publié en ligne: 11 janv. 2023

Pages: 1 - 13

Reçu: 02 déc. 2022

Accepté: 22 déc. 2022

DOI: https://doi.org/10.2478/jfap-2022-0006

Mots clés
Beneish M-score, Artificial neural networks, Predicting auditors’ opinions, Public enterprises

© 2022 Amra Gadžo et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Predicting Auditor’s Opinion on Financial Statements of Public Enterprises Based on Indicators of the Beneish M-score Model

Publié en ligne: 11 janv. 2023

Pages: 1 - 13

Reçu: 02 déc. 2022

Accepté: 22 déc. 2022

DOI: https://doi.org/10.2478/jfap-2022-0006

Mots clésBeneish M-score, Artificial neural networks, Predicting auditors’ opinions, Public enterprises

© 2022 Amra Gadžo et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Mots clés
Beneish M-score, Artificial neural networks, Predicting auditors’ opinions, Public enterprises