Disentangling the black box around CEO and financial information-based accounting fraud detection: machine learning-based evidence from publicly listed U.S. firms
نویسندگان
چکیده
Abstract This study investigates the predictive power of CEO characteristics on accounting fraud utilizing a machine learning approach. Grounded in upper echelons theory, we show value widely neglected for learning-based detection isolation and as part novel combination with raw financial data items. We employ five models well-established literature. Diverging from prior studies, introduce model-agnostic techniques to literature, opening further black box around individual predictors. Specifically, assess predictors concerning their feature importance, functional association, marginal power, interactions. find isolated combined outperform no-skill benchmark approaches by large margins. Nonlinear such Random Forest Extreme Gradient Boosting predominantly linear ones, suggesting more complex relationship between characteristics, data, fraud. Further, Network Size Age contribute second third strongest towards best model’s closely followed Duality. Our results indicate U-shaped, L-shaped, weak L-shaped associations Age, Size, Tenure, fraud, consistent our superior nonlinear models. Lastly, empirical evidence suggests that older CEOs who are not simultaneously serving chairman an extensive network high inventory likely be associated
منابع مشابه
A hybrid model based on machine learning and genetic algorithm for detecting fraud in financial statements
Financial statement fraud has increasingly become a serious problem for business, government, and investors. In fact, this threatens the reliability of capital markets, corporate heads, and even the audit profession. Auditors in particular face their apparent inability to detect large-scale fraud, and there are various ways to identify this problem. In order to identify this problem, the majori...
متن کاملFinancial Accounting Fraud Detection Using Business Intelligence
The paper investigates the inherent problems of financial fraud detection and proposes a forensic accounting framework using business intelligence as a plausible means of addressing them. The paper adopts an empirical case study approach to present how business intelligence could be used effectively in the detection of financial accounting fraud. The proposed forensic accounting framework using...
متن کاملExplain the role of financial knowledge of board members and CEO independence on the quality of accounting information in companies listed on the Tehran Stock Exchange
Among the main factors in creating fundamental changes in the economic environment of Iran, we can name the generalization of ownership of economic enterprises, financing through public participation and privatization of public sectors and economic enterprises. In such circumstances, the transparency and good quality of financial information, which is the basis of optimal economic decisions of ...
متن کاملRelative Earnings and Firm Performance : Evidence from Publicly - listed Firms in China , 2005 - 2012
This paper studies the relationship between three measures of relative earnings and firm performance based on data of 664 listed manufacturing companies in China over the period 2005-2012. It finds that (1) capital earnings relative to labor earnings and the overall average wage level relative to a firm’s average wage level had negative effects on firm performance; (2) the earnings of high-leve...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of Business Economics
سال: 2023
ISSN: ['1861-8928', '0044-2372']
DOI: https://doi.org/10.1007/s11573-023-01136-w