Financial Anomaly Detection: A Six Sigma Approach to Detecting Misleading Financials and Financial Decline

• Activity Number: 24
• Type: Contributed
• Date/Time: Sunday, August 8, 2004 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economics Statistics Section
• Author(s): Radu Neagu*+ and Deniz Senturk and Christina LaComb and Murat Doganaksoy
• Companies: General Electric Global Research Center and General Electric Global Research Center and General Electric Global Research Center and GE Global Research Center
• Keywords: financial ratios ; financial statements ; peer grouping ; z-scores ; six sigma ; financial anomaly

Abstract

Recently, several large companies have collapsed at the top of their game amidst SEC charges of fraudulent financials. Even companies not engaged in fraudulent activities but suffering financial decline are oftentimes undetectable from the handful of financial measures investors and creditors typically examine. With innovative intelligence extraction techniques, holistic and effective insight into a company's financial health can be achieved. We use six sigma tools and extend their applicability to address the problem of profiling a company's financial performance. This way inconsistencies and/or "warning signs" of misleading financial statements are raised and singled-out for separate analysis. The results of applying these techniques on a sample of 22 publicly held U.S. companies prove that our methods give early notice of misleading financial statements and financial behavior many months ahead of events such as SEC investigations or significant drops in company's stock price. We performed a simulation study to understand the dependence and sensitivity of our findings relative to the size of the sample of companies considered in the study and the main results are presented.