Probability Of Fraud Research Articles

Detecting the Probability of Fraud in Interim Financial Statements Using Machine Learning Models: Do Correla-tion-Based Analysis and Principal Component Analysis for Dimensionality Reduction Matter?

Detecting the Probability of Fraud in Interim Financial Statements Using Machine Learning Models: Do Correla-tion-Based Analysis and Principal Component Analysis for Dimensionality Reduction Matter?

An integrated GRC approach to combating fraud in microloan services

This research aims to reduce fraud risks in Indonesian banks and non-bank financial institutions providing microloan services. The study employs data analytics and machine learning techniques using employee data from Bank X spanning 2017-2019 with samples of 28,004 workers (2017-2018) and 27,274 employees (2019). Confirmatory factor analysis and XGBoost predictive modelling are applied within the fraud triangle framework to identify critical fraud risk indicators related to employee pressure. An algorithmic approach categorizes personnel based on fraud risk ratings enabling the detection of potentially suspicious activities for proactive intervention. The analysis reveals that incorporating data analytics into governance, risk management and compliance (GRC) systems can accurately forecast fraud probability by focusing on factors associated with employee pressure and opportunities. This facilitates targeted fraud prevention solutions by integrating control mechanisms, risk processes and auditing standards. The predictive model provides valuable insights for policymakers to combat fraud by enhancing governance and risk management practices specific to microloans. This research concludes that the predictive model is a pragmatic decision-making tool for banks offering micro-loans. It mitigates dangers by detecting high-risk personnel and transactions. Integrating data analytics with robust GRC frameworks enables financial institutions to uphold integrity through proactive fraud monitoring and targeted preventive interventions tailored to identify risk profiles. The study offers an integrated technological organizational approach to protect microlending activities.

The Influence of Corporate Governance Mechanisms on Fraud Probability: Evidence from Russian Companies

This study examines the impact of corporate governance mechanisms on the probability of corporate fraud occurrence. We evaluate the board size, the degree of independence, and the frequency of meetings of the board and its committees. We also attempt to analyse the board’s gender diversity, but since boards are not gender-diverse in Russia, the significance of this variable cannot be tested. Our empirical study is based on 160 observations of MOEX-listed Russian companies, among which fraudulent behaviour has been revealed in 32 companies over a 5-year period from 2014 to 2018. The relationship between the probability of fraud occurrence and corporate governance was investigated employing a logit model. The data was collected from firms’ annual reports and Thomson Reuters Eikon. Data on fraud cases is based on the evidence from the press (including the leading news sources and specialised websites). We detected a significant negative relationship between nomination and remuneration committee chairmen’s independence, the share of independent directors, the independence of board and audit committee chairmen and the likelihood of fraud. We also discovered the insignificant influence of board and its committees’ size and their meetings’ frequency on fraud probability. This paper contributes to the academic research on the relationship between corporate governance mechanisms and probability of fraud occurrence, emphasizing the special role of the establishment of nomination and remuneration committee chairman independence in Russian companies.

Behavioral analysis in forensic setting with the SAVE method

A new theoretical framework developed over the last 10 years, the Behavior Analysis in Medicine, is described to address the challenges faced by forensic and medical or psychological assessment in daily clinical practice. With an interdisciplinary and scientifically based approach that is flexible but systematic in its application, the System for Analysis of Validity in Evaluation (SAVE) is designed to incorporate a wide range of useful methods, techniques and skills drawn from different sources (forensic medicine and psychology, criminology, health law, deception detection, etc.). It is structured in three successive or non-successive phases: verisimilitude (V1), which corresponds to the analysis of the verbal content of the statement; veracity (V2), as the congruence between verbal and non-verbal, and verification (V3), the process of checking the information previously obtained. There is also an overarching phase (V0) or identity assessment, which allows the previous results to be adapted to each case. These knowledge and skills can be learned and trained, all of them have been scientifically tested and have a large number of published references to deepen each phase. In addition, this process allows for easy adaptation to each case and context, avoiding upsetting genuinely impaired examinees and going unnoticed unless the probability of fraud is high. We therefore believe that all forensic, expert witness and legal professionals, especially doctors and psychologists, whether civil servants or independent, will benefit from this way of conducting their assessments, improving the clinical relationship and the necessary management of deception when needed.

Cost-sensitive thresholding over a two-dimensional decision region for fraud detection

Credit fraud poses a challenging task in terms of detection. It can result in significant losses depending on the amount, so a cost-sensitive perspective needs to be taken. Classical approaches focus on estimating the probability of fraud and selecting a decision threshold, but they often fail to consider the transaction amount or account for the cumulative losses incurred within the sample. Consequently, these approaches can result in sub-optimal strategies. A new thresholding approach is proposed, based on the construction of a two-dimensional decision space with an estimated probability and the credit amount. This expansion allows more freedom for the optimal classification rule search, which is performed with a new algorithm. The proposed method generalizes previous approaches, so an improvement is consistently achieved. In addition, it allows a restricted search. This is shown in a study of two real data sets, comparing the results obtained by a wide range of classifiers.

PENGARUH CORPORATE GOVERNANCE TERHADAP CORPORATE FINANCIAL FRAUD PADA PERUSAHAAN MANUFAKTUR YANG TERDAFTAR DI BEI

This study aims to determine the effect of board size, financial expertise, board independence, rationalization, female board, firm size, leverage ratio (LEV) and asset ratio (ROA) on corporate financial reporting in manufacturing companies listed on the Stock Exchange in 2016 - 2020 The sampling technique used is purposive sampling with a final sample of 80 companies. The analysis used in this research is descriptive statistics and panel regression. The results of this study indicate that the size of the board, financial expertise, board independence, rationalization, and do not have a significant effect on the size of the company's finances. While the financial board has a significant influence on fraud, which means a high number of company ownership can reduce the possibility of an incident, then the leverage ratio (LEV) indicates a high debt so there is a high probability of fraud in the company, assets (ROA) has a high positive significant influence ROA can become pressure for the company so that it is likely to take company actions.

An ensemble fraud detection approach for online loans based on application usage patterns

The fraud problem has drastically increased with the rapid growth of online lending. Since loan applications, approvals and disbursements are operated online, deceptive borrowers are prone to conceal or falsify information to maliciously obtain loans, while lenders have difficulty in identifying fraud without direct contacts and lack binding force on customers’ loan performance, which results in the frequent occurrence of fraud events. Therefore, it is significant for financial institutions to apply valuable data and competitive technologies for fraud detection to reduce financial losses from loan scams. This paper combines the advantages of statistical methods and ensemble learning algorithms to design the grouped trees and weighted ensemble algorithm (GTWE), and establishes fraud prediction models for online loans based on mobile application usage behaviors(App behaviors) by logistic regression, extreme gradient boosting (XGBoost), long short-term memory (LSTM) and the GTWE algorithm, respectively. The experimental results show that the fraud prediction model based on the GTWE algorithm achieves outstanding classification effect and stability with satisfactory interpretability. Meanwhile, the fraud probability of customers detected by the fraud prediction model is as high as 84.19%, which indicates that App behaviors have a considerable impact on predicting fraud in online loan application.

A Multi-Module Machine Learning Approach to Detect Tax Fraud

Tax fraud is one of the substantial issues affecting governments around the world. It is defined as the intentional alteration of information provided on a tax return to reduce someone’s tax liability. This is done by either reducing sales or increasing purchases. According to recent studies, governments lose over $500 billion annually due to tax fraud. A loss of this magnitude motivates tax authorities worldwide to implement efficient fraud detection strategies. Most of the work done in tax fraud using machine learning is centered on supervised models. A significant drawback of this approach is that it requires tax returns that have been previously audited, which constitutes a small percentage of the data. Other strategies focus on using unsupervised models that utilize the whole data when they search for patterns, though ignore whether the tax returns are fraudulent or not. Therefore, unsupervised models are limited in their usefulness if they are used independently to detect tax fraud. The work done in this paper focuses on addressing such limitations by proposing a fraud detection framework that utilizes supervised and unsupervised models to exploit the entire set of tax returns. The framework consists of four modules: A supervised module, which utilizes a tree-based model to extract knowledge from the data; an unsupervised module, which calculates anomaly scores; a behavioral module, which assigns a compliance score for each taxpayer; and a prediction module, which utilizes the output of the previous modules to output a probability of fraud for each tax return. We demonstrate the effectiveness of our framework by testing it on existent tax returns provided by the Saudi tax authority.

A Coupled Mathematical Model of the Dissemination Route of Short-Term Fund-Raising Fraud

To effectively protect citizens’ property from the infringement of fund-raising fraud, it is necessary to investigate the dissemination, identification, and causation of fund-raising fraud. In this study, the Susceptible Infected Recovered (SIR) model, Back-Propagation (BP) neural network, Fault tree, and Bayesian network were used to analyze the dissemination, identification, and causation of fund-raising fraud. Firstly, relevant data about fund-raising fraud were collected from residents in the same area via a questionnaire survey. Secondly, the SIR model was used to simulate the dissemination of victims, susceptibles, alerts, and fraud amount; the BP neural network was used to identify the data of financial fraud and change the accuracy of the number analysis of neurons and hidden layers; the fault-tree model and the Bayesian network model were employed to analyze the causation and importance of basic events. Finally, the security measures of fund-raising fraud were simulated by changing the dissemination parameters. The results show that (1) for the spread of the scam, the scale of the victims expands sharply with the increase of the fraud cycle, and the victims of the final fraud cycle account for 12.5% of people in the region; (2) for the source of infection of the scam, the initial recognition rate of fraud by the BP neural network varies from 90.9% to 93.9%; (3) for the victims of the scam, reducing fraud publicity, improving risk awareness, and strengthening fraud supervision can effectively reduce the probability of fraud; and (4) reducing the fraud rate can reduce the number of victims and delay the outbreak time. Improving the alert rate can reduce victims on a large scale. Strengthening supervision can restrict the scale of victims and prolong the duration of fraud.

Managing Ghost Workers Fraud on Payroll in Tanzania: A Conceptual Framework

Tanzania has made several reforms to improve public service since independence in 1961. One of the major concerns has been to wipe out ghost workers on the payroll. The efforts would improve public service and avoid unnecessary wastages of resources. For decades the payroll still bears names of fictitious and ex-employees who either retired or deceased. Ghost workers have been oozing scarce funds that could support service to the neediest Tanzanians. The situation calls for immediate but comprehensive interventions for dealing with menace to avoid further negative impacts on public service management and the country. The core intervention is to comprehend the nature of fraud on the payroll in Tanzania. Therefore, the paper proposes a conceptual framework for identifying and managing ghost workers on the payroll. The study is based on secondary data. The study used Fraud Triangle Theory, Fraud Diamond Theory, and Fraud Lifecycle Management Theory to construct the conceptual framework for managing ghost workers on the payroll in Tanzania. The framework can facilitate organizations, researchers, and human resources practitioners to analyze initial motives and traits of employees towards unethical behaviors, the probability of fraud on the payroll, and the quality of fraud management systems. In addition, the framework portrays that the fraud management systems must be inherent features of efficient and effective human resources management of any organization that needs and wants to curb ghost worker frauds on the payroll.
 Keywords: Ghost Worker Frauds, Fraud Management, Conceptual Framework, Human Capital Management Information System, Payroll

Species substitution and mislabeling in the swordfish (Xiphias gladius) market in Santiago, Chile: Implications in shark conservation

Food fraud and the conservation and trade of threatened species are topics of high concern. Some shark species are considered in this category by IUCN and CITES, and can enter as mislabeled or substituted in the commercialization chain. In Chile, three shark species are present as by-catch in the swordfish (Xiphias gladius) fisheries; these are sold headless, gutted, and finned, hindering their identification by morphological traits. Locally, the commercial designation for swordfish is Albacora, while the unofficial name, Albacorilla, is used in the shark products trade. Thus, DNA barcoding was used for taxonomic identification to assess food fraud incident (FFI) percentage in the swordfish market in Santiago. Muscle samples (n = 47) sold under Albacora or Albacorilla designations were obtained from supermarkets, fish markets, and open-air markets. We predicted the fraud probability by a fitted logistic regression model. The majority (91.48%) of the samples were identified as swordfish, while 6.39% were identified as porbeagle shark (Lamna nasus) and 2.13% as shortfin mako (Isurus oxyrinchus). These shark species are classified as vulnerable and endangered (IUCN), respectively, and listed in Appendix II (CITES). The FFI was 8.52%, including mislabeling (2.13%) found only in fish markets and substitution which was reported in fish markets (4.26%) and open-air markets (2.13%). No FFI were detected in supermarkets. The sales price was significant to explain FFI in the logistic model. We conclude that it is necessary to increase stakeholder awareness about shark conservation status, developing specific labelling legislation containing the scientific names and commercial designations for each shark species, and carry out effective enforcement programs of the labelling regulations using DNA-based methods.

Improving Tax Audit Efficiency Using Machine Learning: The Role of Taxpayer’s Network Data in Fraud Detection

ABSTRACT Using the universe of Armenian business tax payers operating under a standard tax regime, we develop a fraud prediction model based on machine learning tools, with gradient boosting as the primary choice. Having to deal with broadly defined fraud and heterogeneous taxpayers, as well as a relatively small sample, we successfully derive important features from tax returns with a minimum of additional information. Among the important fraud predictors, we obtain historical fraud and audit, share of administrative costs, and external economic activity. We see two main contributions with generalizable practical implications for auditing authorities. First, by focusing on the lift score of the top decile, we demonstrate that even moderately accurate models can improve upon existing accuracy of rule-based approaches. Second, and more importantly, we demonstrate that the information contained in the supplier and buyer network of the taxpayer can be used whenever important predictors of fraud such as historical audits and fraud are not available. This is particularly important for situations with newly established companies, who would otherwise be under-rated in terms of fraud probability.

Segmentation of Potential Fraud Taxpayers and Characterization in Personal Income Tax Using Data Mining Techniques

This paper proposes an analytical framework that combines dimension reduction and data mining techniques to obtain a sample segmentation according to potential fraud probability. In this regard, the purpose of this study is twofold. Firstly, it attempts to determine tax benefits that are more likely to be used by potential fraud taxpayers by means of investigating the Personal Income Tax structure. Secondly, it aims at characterizing through socioeconomic variables the segment profiles of potential fraud taxpayer to offer an audit selection strategy for improving tax compliance and improve tax design. An application to the annual Spanish Personal Income Tax sample designed by the Institute for Fiscal Studies is provided. Results obtained confirm that the combination of data mining techniques proposed offers valuable information to contribute to the study of tax fraud.

The dark side of blockholder control: evidence from financial statement fraud cases

PurposeThe purpose of this study is to investigate whether large blockholders are associated with financial statement fraud at their companies. Although a substantial body of prior studies has focused on chief executive officers’ motivations to manipulate financial statements, the correlation between majority shareholders and financial statement fraud has received little attention. This paper aims to fill this gap by investigating whether the sample firms have controlling shareholders or executives (i.e. blockholders vs management) and whether financial statement fraud schemes, motivations and consequences differ between blockholder- and management-controlled firms.Design/methodology/approachUsing a clinical approach, the authors Study 12 Canadian financial statement fraud cases uncovered by the Ontario Securities Commission between 1997 and 2020.FindingsFirst, the authors find blockholder control in six cases. These findings infer that these large shareholders received private benefits at the expense of minority shareholders. The comparative analyzes suggest that fraudulent firms controlled by blockholders go bankrupt more often than those controlled by managers. The authors also find that improper disclosure is the most common fraud scheme in blockholder-controlled firms.Originality/valueThe authors conduct a deep analysis of financial statement fraud cases to examine the of blockholder control on the likelihood of financial statement fraud. This paper adds new insights to the research on financial crime by investigating whether large shareholders affect the probability of fraud and the extent to which they might do so.

IFDDS: An Anti-fraud Outbound Robot

With the rapid growth of internet finance and e-payment, payment fraud has attracted increasing attention. To prevent customers from being cheated, systems often block risky payments depending on a risk factor. However, this may also inadvertently block cases which are not actually risky. To solve this problem, we present IFDDS, a system that proactively chats with customers through intelligent speech interaction to precisely determine the actual payment risk. Our system adopts imitation learning to learn dialogue policies. In addition, it encompasses a dialogue risk detection module which identifies fraud probability every turn based on the dialogue state. We create a web-based user interface which simulates a practical voice-based dialogue system.

Identification and Prediction Methods of Financial Anti-fraud

Nowadays, evaluating and identifying the potential fraud risk of borrowers effectively and calculating the fraud probability of them are the basis and significant steps of credit risk management in modern financial institutions before issuing loans. This paper mainly studies the statical analysis of the historical loan data of financial institutions based on the idea of unbalanced data classification and establishes the prediction model of loan fraud through random forest, decision tree and regression algorithm. The prediction performance of random forest algorithm is better than the other two mentioned methods. Additionally, it may obtain the feature that have a remarkable impact on the final fraud by ranking the importance of those features, which leads to a more effective judgment on the credit risk in the financial field.

Ultimate Control Rights and Corporate Fraud: Evidence from China

ABSTRACT This study uses the Shapley-Shubik Power Index to measure the ultimate control rights of companies listed on China’s main board market and empirically tests the effect of ultimate control rights on corporate fraud. We confirm that ultimate control rights have a significant restraining effect on corporate fraud. Specifically, the greater the ultimate control rights, the lower the probability of fraud and the fewer the number of frauds. Furthermore, we find that under different fraud subjects, fraud types, and fraud severity, this restraining effect still exists. In addition, we also find the ultimate control rights are negatively associated with fraud propensity, and are positively associated with fraud detection.

Do financial analysts discourage or encourage corporate fraud? Empirical evidence from China

Purpose This paper aims to focus on the role of financial analysts in corporate fraud in the Chinese stock market. Design/methodology/approach Data on the analyst coverage and all the types of corporate fraud were obtained for 16,284 company-year observations of Chinese companies. The sample was subsequently divided into those of state-owned enterprises, before and after financial crisis. Findings The overall results indicate that analyst coverage effectively deters the occurrence of fraud. The sub-sample results suggest that the impact of analysts on deterring fraud is more pronounced in non-state-owned enterprises, especially after the financial crisis. The path analyses show that analyst coverage can deter corporate frauds by affecting information transparency and investor attention. Furthermore, the results show that the deterrence role of financial analysts varies with fraud types: it is more pronounced in deterring disclosure fraud, but not as effective in illegal guarantees and illegal insider dealing. Moreover, analyst coverage can deter the occurrence of fictitious reporting, intentional postponement and material omission. Originality/value This paper not only examined the overall fraud probability but also taking into consideration the heterogeneity of the information availability and research focus of financial analysts and examined the analysts’ impact on the occurrence of difference types of fraud. Moreover, this paper explored why financial analysts can deter corporate frauds through path analyses.

Insider Abuse and Fraud Prediction for U.S. Banks: A Comparison of Machine Learning Approaches

We develop and compare alternative empirical models for the detection of insider abuse and fraud occurring at U.S. commercial banks with the goals of identifying leading indicators of fraud and fraud prediction. Specifically, we use information on enforcement actions taken by bank supervisors to remove bank officers and other employees and to prohibit bank employees from banking during the period 2007–2019 to identify instances of serious insider abuse and fraud. To the best of our knowledge, these data — enforcement actions — have not been used in previous studies of insider abuse and fraud (hereafter, fraud) at banks. The explanatory variables in our models are measures of banks’ capital adequacy, asset quality, management, earnings, liquidity and sensitivity to market risk and other financial measures that previous studies have identified as indicators of fraud. We model fraud using machine learning approaches, each with different strengths and weaknesses — artificial neural network, deep learning autoencoder network, gradient boosting, logit regression and random forest. The models are fitted annually using observed incidences of fraud and four quarters of bank financial data. We use these models to make predictions of the probability of fraud at banks in the following calendar year. Our results indicate that artificial neural network, gradient boosting and random forest are, in general, more accurate in out-of-sample/time fraud forecasts than are deep learning autoencoder network and logit regression models. Finally, in terms of the drivers of fraud the estimates of logit regressions are somewhat mixed but do suggest that earning assets and capitalization are negatively related to fraud while loan loss rates and noncore funding dependence are positively related to fraud.

Corporate Social Responsibility and Financial Fraud: The Moderating Effects of Governance and Religiosity

This study investigates how managers in firms that have committed fraud strategically use socially responsible activities in coordination with their fraudulent financial reporting practices. Using propensity score matching to select control firms that have a similar probability of fraud in the pre-fraud benchmark period, we find that the corporate social responsibility (CSR) performance of fraudulent firms in the fraud-committing period is significantly higher compared with the CSR performance of non-fraudulent control firms during this period, and compared with that during their own pre-fraud benchmark periods. This higher CSR performance by fraudulent firms is achieved by means of investing in both stakeholder and third-party CSR categories and by improving in CSR strengths. Furthermore, the increase in CSR performance is more pronounced for fraudulent firms with a weak governance environment, and for firms located in high-religiosity states. Overall, our findings suggest that fraudulent firms strategically adjust their CSR performance to coordinate with their fraudulent financial activities.