UVA-BYU MetaFraud: SEC droid we’re looking for?

by Kurt Schulzke

Researchers at University of Virginia and BYU have developed what may be the Holy Grail of securities-fraud surveillance: a software algorithm they’ve named MetaFraud. In a forthcoming article in MIS Quarterly, co-authors Ahmed Abbasi, Conan Albrecht, Anthony Vance, and James Hansen, describe the algorithm, which they say identifies financial statement fraud with 80-percent accuracy. According to a BYU news release,

The researchers funneled 9,000 instances (instance = financial information for one firm for one year) from more than 15 years through MetaFraud, and compared its predictions of fraud with known fraudulent firms identified by the SEC. MetaFraud correctly predicts fraud with 80 percent accuracy, and over 90 percent accuracy when MetaFraud reports a high level of confidence, which it was able to do 70 percent of the time.

More from co-author Tony Vance:

Sounds great.  Of course, as good as MetaFraud sounds, no software program, no matter how reliable, can equal the precision of boots-on-the-ground whistleblower intelligence.  MetaFraud, however, is an encouraging development that may help outsider SEC fraud monitors focus their attention on the most likely high-value targets. With MetaFraud now in the can, if Abbasi & Co. are taking requests, perhaps they could whip up a Medicare adaptation: MediFraud?

Previous post:

Next post: