Kx Insights: Time for a new approach to financial surveillance

3 Jul 2018 | , ,
Share on:

By Michael Gorman

Kx and First Derivatives regulatory expert Michael Gorman frequently addresses current issues facing legal, credit and risk oversight personnel at financial institutions on the Kx blog. In this latest article he takes a look at the growing importance of a next generation approach to surveillance systems for financial services companies.

Is your financial services company prepared for increased enforcement? In 2017, for example, the US Commodities Futures Trading Commission (CFTC) tripled the number of enforcement actions involving disruptive trading practices, levying tens of millions of dollars in fines and permanently barring traders from futures and options markets regulated by the CFTC. Though they don’t agree on many things, global regulators are united in upping the appetite for enforcement. If you got that ‘notice of investigation,’ how long would it take you to reconstruct the trade?

Financial services compliance and business operations teams don’t often have a free moment for introspection. From implementing regulatory changes, to responding to regulators’ increasing appetite for enforcement, most teams have to be reactive — providing reports and constantly revising their processes. Yet, as difficult as finding pensive time is, a few moments of pause to take stock are worth the effort—and can help shape future enhanced outcomes at lower cost of ownership.

Here are a few points to consider, and to use to measure the state of your own firm’s approach to surveillance:

Are you only seeing a small corner of the universe?
Competing regimes in competing jurisdictions have created siloed data and systems (whether by geography or asset class or regime) that obscure the enterprise view and make surveillance more difficult.

Can your personnel manage the workload?
Given the breadth of regulation that now exists, surveillance teams are struggling to provide the necessary coverage to monitor and clear an increasing volume of alerts. Even incremental additions to the business and surveillance teams are rarely effective in keeping up with the pace, with growing lists of suspicious events being assigned to overburdened analysts.

Are you efficiently managing high alert volume?
Prioritizing alerts and maximizing staff responsiveness to alerts means both using alert engines that self-learn and using intuitive visualizations that will allow teams to focus on what matters.

Can your firm bridge the compliance/business divide?
Regulators are now ‘encouraging’ firms to shift the first line of defense to business operations. How does that look in your firm? How should responsibilities be divvied up? More importantly, how do first line compliance activities work day-to-day?

For most, no clear answers present themselves, but a clear challenge does. Firms need to understand how to keep their compliance function independent and encourage cooperation from the business side, so both groups assist each other in investigating market abuse events.

Today, lines of inquiry, and the communication of evidence, between these groups is usually done through antiquated formats and is a real inhibitor to effectively flagging and managing market abuse cases. Both sides are crying out for a common platform that has adequate access controls and tools that help them liaise on mutual areas of concern.

Does your system allow you to remain technology capable?
Having the right surveillance staff and domain knowledge is the first step, but if they feel that their data management infrastructure is not up-to-date, they will feel they are working at a disadvantage, especially if the other side is using more powerful weapons. Agile data management, with the ability to process and analyze data at speed, is at the core of a responsive surveillance system.

Legacy systems in most firms typically provide user interfaces that allow analysts to manage investigative workflows but give little thought to the processing power or data granularity required to create useful outputs. Marrying these key technological ingredients in a cost-effective single platform is essential to staying on par with the most sophisticated market participants.

Are your responses purely tactical?
Tight deadlines have necessarily perpetuated the bad habit of tacking on quick fixes to legacy systems, increasing risk exposure and inhibiting your firm’s ability to manage alerts thoroughly. With new technology for surveillance coming to market, now is the time to step back and think about a system-wide upgrade.

Are you building for the future?
Machine learning is the future, and we will all get there, but expectations have far outstripped current capabilities for predictive surveillance. When that time comes, the effectiveness of any machine learning technique will depend on a platform that can feed the machine the data it needs to learn. Data really is the new oil and machine learning will require a lot of diesel.

The new approach to surveillance
Keeping up with the greater frequency of regulatory rule changes and higher demands for timely reporting by regulators demands a new approach to financial surveillance. Legacy systems simply are not capable of supporting today’s requirements.

Modern surveillance systems, like from Kx, need to instantly detect known trading violations like layering, spoofing or marking the close. Customers need to be able to calibrate their parameters in real time to improve their detection quality and accuracy. Plus their historical database and replay engine needs to be easily adjustable for retrospective investigations into new types of fraudulent behavior and suspicious activity.

If you are interested in finding out more about today’s approach for financial surveillance, and how to future-proof your systems, contact us at .

Michael Gorman is Global Head of the Regulatory and Compliance practice at Kx parent First Derivatives plc. His team combines subject-matter expertise, strong product knowledge, and data and technology DNA to offer a range of consulting services that can accelerate implementation of regulatory solutions. Read more here.

SUGGESTED ARTICLES

ML and kdb+

Machine Learning Toolkit Update: Cross-Validation and ML Workflow in kdb+

23 Jul 2019 | , ,

The Kx machine learning team has an ongoing project of periodically releasing useful machine learning libraries and notebooks for kdb+. This release relates to the areas of cross-validation and standardized code distribution procedures for incorporating both Python and q distribution. Such procedures are used in feature creation through the FRESH algorithm and cross-validation within kdb+/q.

kx and machine learning

Machine Learning Toolkit Update: Multi-parameter FRESH and updated utilities

25 Apr 2019 | , ,

This latest toolkit release, is the first in a series of planned releases in 2019 that will add updates to the functionality of the FRESH (Feature Extraction based on Scalable Hypothesis tests) algorithm and the addition of a number of accuracy metrics, preprocessing functions and utilities. In conjunction with code changes, modifications to the namespace structure of the toolkit have been made to streamline the code and improve user experience.