Predicting accounting fraud using imbalanced ensemble learning classifiers – evidence from China

Abstract

The current research aims to launch effective accounting fraud detection models using imbalanced ensemble learning algorithms for China A‐Share listed firms. Based on a sample of 33,544 Chinese firm‐year instances from 1998 to 2017, this research respectively established one logistic regression and four ensemble learning classifiers (AdaBoost, XGBoost, CUSBoost, and RUSBoost) by 12 financial ratios and 28 raw financial data. Additionally, we divided the sample into the train and test observations to evaluate the classifiers' out‐of‐sample performance. In detail, we applied two metrics, namely, Area under the ROC (receiver operating characteristic) curve (AUC) and Area under the Precision‐Recall curve (AUPR), to evaluate classifiers' discriminability. In the supplement test, this study put forward an algebraic fused model on the basis of the four ensemble learning classifiers and introduced the sliding window technique. The empirical results showed that the ensemble learning classifiers can detect accounting fraud for the imbalanced China A‐listed firms far more effectively than the logistic regression model. Moreover, imbalanced ensemble learning classifiers (CUSBoost and RUSBoost) effectively performed better than the common ensemble learning models (AdaBoost and XGBoost) in average. The algebraic fused model in the supplement test also obtained the highest average AUC and AUPR among all the employed algorithms. Our results offer firm support for the potential role of Machine Learning (ML)‐based Artificial Intelligence (AI) approaches in reliably predicting accounting fraud with high accuracy. Similarly, for the Chinese settings, our ML‐based AI offers utmost advantage in forecasting accounting fraud. Finally, this paper fills the research gap on the applications of imbalanced ensemble learning in accounting fraud detection for Chinese listed firms.

An intelligent detecting model for financial frauds in Chinese A‐share market

“…We further identify a list of 28 raw financial data items from corporation financial statements which are used to calculate above selected 12 financial indicators, following Sun et al (2021) and Rahman & Zhu (2023). As shown in Table 2, there are 21 items from the Balance Sheet (i.e., Current assets; Cash; Short‐term investments; Current liabilities; Debt in Current Liabilities; Total assets; Available‐for‐sale securities; Held‐to‐maturity securities; Long term equity securities; Total liabilities; Long‐term debt; Debt in current liabilities; Accounts receivables; Inventory; Property, plant and equipment; Construction in progress; Project material; Accounts payable; Surplus reserve; Retained earnings; Income taxes payable ), five items from the Income Statement (i.e., Sales; Cost of goods sold; Net income; Financial expenses; Income taxes ), and two items from the Cash flow statement (i.e., Cash received from absorbing investments; Proceeds from issuance of bonds ).…”

“…We apply two metrics, AUC and NDCG@k, to evaluate different models' fraud detection performance. These two metrics are widely applied to assess the performance of classifiers (Sun et al, 2021; Rahman & Zhu, 2023; Li et al, 2023).…”

“…Fortunately, machine learning develops rapidly in recent years, providing efficient approaches to exploring the relationship between financial risks and the growing financial data (Du & Shu, 2022;Wu et al, 2022). Therefore, many scholars are devoted to developing novel fraud detection models using machine learning, such as Logistic Regression, Naive Bayes, Support Vector Machine, Neural Network, Random Forest, Ensemble Method and many more (Song et al, 2014;Cao et al, 2015;Vasarhelyi et al, 2015;Brown et al, 2020;Ding et al, 2020;Bertomeu et al, 2021;Chen & Zhai, 2023;Xu et al, 2023;Achakzai & Peng, 2023;Pan et al, 2023;Riskiyadi, 2023;Rahman & Zhu, 2023;Zhou et al, 2023).…”

CADM: Big Data to Limit Creative Accounting in Saudi-Listed Companies

“…ML and DL models can be used parallelly to build a model with improved capabilities, as in Hybrid Learning (HL). They can also be combined, and the output of one model is the input for the second, and that is called Ensembled Learning (EL) as in [49] [29]. Accordingly, the training process of CADM uses an HL approach for enhanced performance.…”

“…Other N-FIN features can be included according to their availability in Saudi sources, such as board meetings, meeting minutes, and meeting content, as tested in [26]. As shown in TABLE IV., FIN and N-FIN features selection will follow the criteria in [50], [49], and [51] accordingly. For any AI model to be accurate, variables should be limited [52], though this study proposes to test extended obtainable variables and exclude some in the exploration process if needed.…”