Predictive Analytics Can Defeat Rogue and Insider Trading

It seems there is no end to new cases of market manipulation, as the latest scandal involving banks possibly rigging the $1.5 trillion government-sponsored bond market breaks in Europe.

Regulators have been relentless in cracking down on financial services firms involved in insider trading and market manipulation, and financial services firms are going to great lengths to comply with new regulations. Yet, the scandals continue.

Therefore, this year we will begin to see wider usage of predictive analytics in capital markets in order to really crack down on any kind of aberrant activity before it happens.  

Increasingly, regulators and financial services firms will monitor trades and traders to spot unusual behavior and anomalous trades, using streaming analytics with predictive analytics models on top. This way they will be able to predict with a good deal of certainty when something bad might happen; insider trading, market manipulation, even money laundering can be halted before markets are affected.

Converging siloed systems such as anti-money laundering, operational risk and trader profiling into a single monitoring system enables a more correlated view of potential threats. Adding real-time information to historical data gives firms the ability to perform “continuous analytics,” allowing the extrapolation of what has happened so far to predict that something might be about to happen—and prevent it.

Including data from social media, email and chat rooms helps alert management to anomalous behaviour, such as traders working longer hours than usual or deals with unusual counterparties.

Surveillance across asset classes goes a long way to preventing rogue algorithms from running amok in more than one market. Cross-region monitoring assists compliance managers in assuring adherence to different regulatory environments.

Monitoring for “unknown unknowns,” by benchmarking behavior that is “normal” over time and spotting behavior that deviates from the norm, helps prevent fraudulent behavior.

Turning a new unknown behavior into a known behavior means adding new rules to the system, giving capital markets firms the power to evolve dynamically as regulations change.

Using predictive analytics, firms can detect unwanted behaviors as they happen, even in micro milliseconds as with HFT, such as system errors, insider trading, front-running, wash trading or quote stuffing.

They can then act on them, thus preventing trade-related meltdowns such as the London Whale, whose shenanigans cost JP Morgan $6 billion,  or Knight Capital, where a rogue algorithm cost it nearly half a billion dollars. Or spotting a cartel manipulating bond prices.