A Robust Multivariate Outlier Detection Method for Detection of Securities Fraud

A Robust Multivariate Outlier Detection Method for Detection of Securities Fraud

M. Fevzi Esen
Copyright: © 2020 |Pages: 18
DOI: 10.4018/IJBAN.2020070102
(Individual Articles)
No Current Special Offers


Insider trading is one the most common deceptive trading practice in securities markets. Data mining appears as an effective approach to tackle the problems in fraud detection with high accuracy. In this study, the authors aim to detect outlying insider transactions depending on the variables affecting insider trading profitability. 1,241,603 sales and purchases of insiders, which range from 2010 to 2017, are analyzed by using classical and robust outlier detection methods. They computed robust distance scores based on minimum volume ellipsoid, Stahel-Donoho, and fast minimum covariance determinant estimators. To investigate the outlying observations that are likely to be fraudulent, they employ event study analysis to measure abnormal returns of outlying transactions. The results are compared to the abnormal returns of non-outlying transactions. They find that outlying transactions gain higher abnormal returns than transactions that are not flagged as outliers. Business intelligence and analytics may be a useful strategy for detecting and preventing of financial fraud for companies.
Article Preview

1. Introduction

Providing timely, reliable and precise flow of information is an essential factor for financial markets that creates equal opportunity and strengthens investors' decisions and views on the market. In strong form of market efficiency, it is assumed that all of the available information regarding the securities is equally distributed to all investors and no one can profit more than other investors. Prices are assumed to reflect not only the public information but also the private information, which is only available to insiders who access inside information. However, markets were found to be semi-strong and weak form efficiency in empirical studies, which have evidence against market efficiency (Thaler, 1999; Malkiel, 2003). These studies show that markets cannot achieve efficiency even in semi-strong form and investors can make abnormal returns.

The world economy is suffering an average loss of about $4.39 trillion every year due to deliberate concealment of misinterpreted financial statements and unfair, deceptive practices (Crowe Clark Whitehill, 2017). As a consequence of financial crimes, one third of companies are in a financial turmoil and more than half of these companies experience losses of 100 thousand dollars or more. According to global economic crime survey of 2016, illegal insider trading is regarded as one of the three major risks affecting stock markets and it is shown as the fastest growing economic crime by 75% in 2016 compared to the previous year (PwC, 2016).

One in ten people in financial services industry is suffering a financial loss of five million dollars or more after a fraud. In stock exchanges, for every $100.000 transaction, the estimated cost of illegal insider trading is about $1 per investor (Ali and Gregoriou, 2008). The total cost of illegal insider trading may be clearly understood when indirect costs such as loss of productivity, operating expenses, lack of motivation, and volatile stock prices are involved in estimated direct economic losses.

Illegal insider trading is one of the main financial crimes, refers to trading with material non-public inside information that reduces information transparency and violates investors' confidence in the market. For this reason, it is aimed to prohibit the use of inside information and prevent the monopoly of information by conveying of inside information to the all investors in a timely manner.

The increasing number of transactions in financial markets make the use of quantitative and qualitative data necessary to detect anomalies and abnormal gains. Due to continuous refinement of the methods of fraudsters and various case based characteristics of financial frauds, many data mining approaches have been proposed accordingly. In order to ensure that managers develop suitable strategies to mitigate the effects of financial frauds, there is a requirement for understanding the dynamics of financial transactions and fraud detection strategies together.

Only in NYSE, the average daily trading volume between 2010 and 2017, is estimated to be about 1.25 billion shares with a total market value of $41.89 billions. This shows the necessity of information discovery from databases, namely, the use of data mining, which harbours probability, statistics and machine learning techniques. Data mining can assess financial crime detection problems in a dynamic and task-oriented framework and it is able to model past and present patterns of data (Cahill et al., 2002).

This paper contributes to the fraud detection on financial markets literature. Our research on financial fraud detectiton shows that there are very few papers published on data mining applications for insider trading surveillance in stock markets. None of the studies examining insider trading have detected unusual insider transactions by using data mining techniques and subsequently calculate abnormal returns of insider transactions. This study is an effort for further examination of insider transactions and the variables that affect insider trading profitability.

Complete Article List

Search this Journal:
Volume 11: 1 Issue (2024)
Volume 10: 1 Issue (2023)
Volume 9: 6 Issues (2022): 4 Released, 2 Forthcoming
Volume 8: 4 Issues (2021)
Volume 7: 4 Issues (2020)
Volume 6: 4 Issues (2019)
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 4 Issues (2015)
Volume 1: 4 Issues (2014)
View Complete Journal Contents Listing