Financial Fraud Identification of the Companies Based on the Logistic Regression Model.

Companies’ financial fraud provokes declining market asset allocation efficiency and significantly impacts trust and loyalty among all company’s stakeholders. Most investigations focused on the prediction of accounting fraud; less research concentrated on financial restatements. In this case, the paper aims to develop a model for identifying the companies’ financial fraud according to the developed index system construction based on financial statements and relationships between their items. The study applies the following stages: 1) analysis of the theoretical framework of the core determinants, impulses and factors of financial fraud and their identification; 2) development of the methodology for timely identification of financial fraud, which is based on index system construction using the Logistic regression model. The object of investigation is Chinese companies listed by China Stock Market & Accounting Research database, excluding J66 (remaining financial industry except for the monetary and financial services), J67 (capital market services), J68 (insurance industry) and J69 (other financial industry) enterprises. The period of investigation is 2017–2020. The data sample includes 53 fraudulent and 53 normal Chinese enterprises. The results show that the overall prediction accuracy of the developed model is 83% and robustness test results further verify the rationality and effectiveness of the method. The company’s stakeholders could apply the proposed approach for fraud identification to improve the efficiency of financial fraud identification from the technical level. [ABSTRACT FROM AUTHOR]