TradeTech: AI/ML use split across ‘haves’ and ‘have nots’

There is a split on deployment artificial intelligence (AI) and machine learning (ML) use on trading, with 52% deploying these technologies in their trading processes, according to a poll taken at a Trade Tech data and analytics panel discussion on the subject.

The survey found that the most common use was to enhance execution which came in at 38% while only 10% leverage them for analytics and a mere 5% for stock selection.

Camillia Zedan, Morgan Stanley.

Instead, many buy and sell side firms are looking more carefully at some of the tried and tested tools that are already are fixtures in the marketplace. “At the moment, we are seeing people being much more forensic about where they send orders and are focusing more on transaction cost analysis instead of these more sophisticated technologies,” says Camillia Zedan – EU head of execution analytics, Morgan Stanley’s Electronic Trading (MSET c Trading (MSET), one of the panellists on How can you successfully apply AI/ ML and advanced analytics techniques to enhance your trading performance in constantly changing market conditions?

Antish Manna, MAN Group.

Fellow panellist Antish Manna head of trading analytics, Man Group also believes the use of these techniques will depend on the size of the organization and their order book. “AI/ML are not applicable to all strategies,” he says. “They are the most useful if a firm has a huge sample – millions of orders worth billions of dollars – because these technologies will be able to provide the rich data sets insights that traders will need.”

Although change always takes time, there has though been an acceptance that new technologies will not completely take over the financial services world. In the past there was a view that stronghold legacy systems had on buy and sell side firms would be undone but today, their grip is expected to remain. The buzzwords now are interoperability and open source.

These were some of the themes debated at the sister panel – How can you build a centralised data framework and technology stack that breaks down high volumes of data and streamlines your workflow to make intelligent trading decisions.

Dan Schelifer.

“Legacy systems are not going away,” says panellist Dan Schleifer, CEO, Cosaic. “For example, in the past we all thought that all the data would go into a data lake, and it would be a seamless user experience but that will never happen. The focus today is how we can improve the experience and pull all the data from the silos and workflows across the applications that are contextually relevant and minimize the desktop surfing.”

Natixis Investment Management has already tackled some of these issues. “There was a proliferation of data in traditional silos, and we built a new platform that allows us to look at the traditional as well as signals from social media and other unstructured source,” says panellist George Marootian, EVP and head of technology at Natixis Investment Managers. “This enables us to have a free flow of consistent, reliable and portable data across the organisation.”

 

Unblocking the carbon credits bottleneck

Claire Dorrian, head of sustainable finance, capital markets at London Stock Exchange Group (LSEG),...

Deutsche Bank tightens vendor sustainability requirements

Deutsche Bank is tightening its sustainability requirements for vendors and suppliers across the world and...

London Stock Exchange Group acquires MayStreet

London Stock Exchange Group (LSEG) has signed an agreement to acquire MayStreet, a US-based...

Unblocking the carbon credits bottleneck

Claire Dorrian, head of sustainable finance, capital markets at London Stock Exchange Group (LSEG),...

Deutsche Bank tightens vendor sustainability requirements

Deutsche Bank is tightening its sustainability requirements for vendors and suppliers across the world and...

London Stock Exchange Group acquires MayStreet

London Stock Exchange Group (LSEG) has signed an agreement to acquire MayStreet, a US-based...