Best financial crime product and Best overall provider of the year: Nasdaq Smarts

A surveillance system developed by Nasdaq proves critical to the success of a 'layering' case

nasdaq-michael-o-brien
Michael O'Brien, head of product management for Smarts Trade Surveillance at Nasdaq

OpRisk Awards 2016

In August 2015, the UK's Financial Conduct Authority scored a major victory in its fight against market abuse, when the High Court imposed permanent injunctions against two investment firms – Da Vinci Invest and Mineworld Limited – and three Hungarian traders, fining them £7.57 million ($11 million) for market manipulation that had been identified as far back as 2010.

The particular strategy for which they were prosecuted is known as ‘layering', whereby the order book is manipulated to create a misleading impression of supply and demand, so that shares can be traded at artificial prices. It has historically been a difficult crime to prove, given the multiple factors involved and the need to demonstrate intent, but a surveillance system developed by Nasdaq proved critical to the success of this particular case.

"Our market replay visualisations were used to show the judge and jury exactly how the market had been manipulated. This was one of the most significant layering cases to have been prosecuted, and it showed clearly how layering scenarios can be constructed," says Michael O'Brien, head of product management for Smarts Trade Surveillance at the New York-based exchange and technology vendor.

The Smarts platform has existed for more than 20 years, but has successfully kept pace with the evolution of the industry. Targeted initially at exchanges and regulators, it has extended over the past decade and the platform's client base now includes 119 sell- and buy-side firms, as well as 45 exchanges and trading platforms, and 13 regulators.

Our market replay visualisations were used to show the judge and jury how the market had been manipulated. This was one of the most significant layering cases to have been prosecuted, and it showed how layering scenarios can be constructed
Michael O'Brien, Smarts Trade Surveillance

As for any surveillance system, the greatest challenge is to sift through masses of data and generate timely alerts relating to suspicious activity without bombarding the user with unnecessary messages or neglecting relevant information. Enhancing the platform's information and alert management has been a significant area of investment for the business.

"The platform needs to cut through millions of messages and identify incidents that are truly unusual. We are increasingly taking in contextual information and external triggers, such as news reports, historic activity, market data and events, e-comms data and profitability levels, to make sure the correct scenarios are flagged," says Valerie Bannert-Thurner, head of risk and surveillance solutions at Nasdaq.

Combining information

The problem is that many large institutions tend to work in siloes, making it particularly difficult to combine information from across the business and reconstruct trading scenarios accurately. Effective surveillance requires technology that is equipped not just with transactional pattern recognition, but also with language processing and machine intelligence, according to Bannert-Thurner.

That realisation led the Smarts team to evaluate a number of technology providers. In February, it announced a strategic alliance with Digital Reasoning, a Tennessee-based firm specialising in cognitive computing. By integrating its trade surveillance with natural-language processing and machine intelligence-based technology, Nasdaq believes it can enable Smarts users to better understand the intent and meaning behind particular patterns of behaviour.

"This is a very unique partnership where we have distribution rights to Digital Reasoning's e-comms monitoring technology in the space of surveillance and compliance towards capital markets, which will allow us to strengthen our offering and deliver a truly holistic approach to surveillance," says Bannert-Thurner.

  • LinkedIn  
  • Save this article
  • Print this page  

You must be signed in to use this feature.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: