Predicting the manipulation of financial statements of Tehran Stock Exchange companies using the Benish model and Bayesian networks

The increase in manipulation of information and financial statements of companies, as well as the occurrence of fraud and restatement of financial statements, which often lead to the distress and bankruptcy of companies, has raised concerns about the quality of information in financial statements. Given the importance of this issue, discovering or predicting the occurrence of these manipulations and the factors affecting them has always been of interest to researchers, analysts, investors, and managers in companies. Therefore, the purpose of this research is to predict the manipulation of financial statements of companies listed on the Tehran Stock Exchange using the Benish model and the Bayesian network model, as well as to compare the performance of these models in predicting the manipulation of financial statements with each other. This research is applied in terms of purpose, quantitative and post-event in terms of data, and descriptive-correlation in terms of analysis. The statistical population of the research was all companies listed on the Tehran Stock Exchange in the period ۲۰۱۸ to ۲۰۲۲, and the samples were selected using the systematic elimination method. The criterion for selecting companies with financial statement manipulation was that the companies had an unqualified audit opinion with a qualified clause subject to distortion in financial data or the existence of tax disputes with the tax authority according to the income tax reserve note and tax file and the conditional clause of the audit report or the existence of significant annual adjustments and restated financial statements. The research data was collected using library and document mining methods and analyzed using EViews software. The results showed that the Benish model, with an accuracy of ۸۴.۲۶\% and Bayesian networks, with an accuracy of ۹۰\% have the ability to predict financial statement manipulation among companies listed on the Tehran Stock Exchange. Also, according to the research results, the performance of Bayesian networks, which are artificial intelligence models, in predicting financial statement manipulation is better than the performance of the Benish model, which is a linear model.

کلیدواژه ها:

Financial statement manipulation ، Benish model ، Bayesian network model