Computing Community Consortium Blog

The goal of the Computing Community Consortium (CCC) is to catalyze the computing research community to debate longer range, more audacious research challenges; to build consensus around research visions; to evolve the most promising visions toward clearly defined initiatives; and to work with the funding organizations to move challenges and visions toward funding initiatives. The purpose of this blog is to provide a more immediate, online mechanism for dissemination of visioning concepts and community discussion/debate about them.

Pinpointing Anomalies in Complex Financial Data

October 31st, 2011 / in Research News / by Erwin Gianchandani

Battelle has licensed its Anomalator software, which pinpoints anomalous information in financial data to identify problematic trends that can put the global financial system at risk [image courtesy Battelle via PNNL].Researchers at the Department of Energy’s Pacific Northwest National Laboratory (PNNL) have developed new software that helps identify anomalies in complex financial data, in hopes of detecting problematic financial trends that jeopardize U.S. and global financial systems. It’s a great example of the kinds of research opportunities at the intersection of computer science and finance.

From the official DoE press release:

Identifying atypical information in financial data early could help identify problematic financial trends such as the systemic risk that recently put the U.S. and global financial systems in a downward fall. Recognizing such anomalous information can also help regulators, investors and advisors better manage their investment and savings portfolios.


Now, new analytical software developed by Battelle researchers based [at PNNL] can do just that…


In a demonstration of this technology, the Battelle-developed Anomalator™ software recently picked out the atypically stable and positive returns reported by disgraced financier Bernard Madoff as an anomaly among hundreds of funds. He is now serving a 150-year prison sentence for scamming investors out of as much as $65 billion in a Ponzi scheme that spanned at least 20 years. Unfortunately, Madoff’s fraud was concealed for more than 15 years. The use of this sophisticated anomaly-detection and visualization tool could have exposed Madoff early on, and can help expose future scandals, its inventors say…


Traditional financial analysis reports either provide a list of numbers or a simple line graph to represent the value of just one investment over time. The Anomalator™ is unique in its ability to identify unusual trends in complex financial data and graphically show how it compares with larger datasets.

So how does it work? Find out after the jump…

Anomalator™ uses mathematical algorithms to identify the atypical data in databases that record the movement of funds or the people who manage them over time. The software then creates a line graph representing the progress of anomalous funds or managers, as well as other user-selected funds or managers of interest…

In the case of Madoff:

Homing in on the tool’s potential to detect financial fraud, [researchers] compiled Madoff’s stated returns from one of the leading Madoff feeder funds… ran several scenarios and found that while the majority of the market was volatile, repeatedly spiking up or down, Madoff’s returns were atypically upward sloping and effectively never lost money. This technology’s unique anomaly-detection and visualization helps expose glaringly anomalous patterns such as those produced by Madoff.

To learn more, check out the DoE press release.

(Contributed by Erwin Gianchandani, CCC Director)

Pinpointing Anomalies in Complex Financial Data

Comments are closed.