TECH TUESDAY is a weekly content series covering all aspects of capital markets technology. TECH TUESDAY is produced in collaboration with Nasdaq.
The latest iterations of artificial intelligence (AI) have lots of buzz, but how is the technology changing trading in capital markets?
That bottom-line consideration was the topic of Nasdaq’s inaugural Market Makers webinar, held on July 24. The one-hour event was hosted by Nasdaq Chief Economist Phil Mackintosh and featured Doug Hamilton, Nasdaq’s Head of AI Research and Engineering, and Markus Pelger, Assistant Professor of Management Science and Engineering at Stanford University and a faculty research fellow at the National Bureau of Economic Research.
Hamilton covered how Nasdaq is leveraging AI to modernize markets, in particular, to create a more dynamic capital market infrastructure.
He opened his remarks by noting that today’s capital markets are highly efficient, but there are inefficiencies in core functions, such as liquidity provisioning. “How can we create and design markets to bring liquidity?” Hamilton asked.
Nasdaq approaches AI from two perspectives: the technology itself, and the business value of the technology. One early conclusion reached by the exchange operator was that one-size-fits-all doesn’t work for all AI; bespoke solutions are often needed.
One bespoke AI solution is Nasdaq’s Dynamic Midpoint Extended Life Order (M-ELO) , in which an AI-powered timer seeks to simultaneously improve both liquidity and execution quality outcomes – usually thought to be in tradeoff – by adjusting based on observed market behavior.
Hamilton next walked through how Nasdaq is applying AI in listed options by managing strike prices in a sprawling market with more than 1.4 million options contracts. “We developed a system that predicts volume and makes a more optimal list of strikes on an exchange,” he said.
He added that the end result has been “phenomenal,” with 30 million more options contracts traded in 2023 as a result of better-aligning strikes with market demand. “It allows people to trade what they want to trade,” he said.
Hamilton summed up his presentation, concluding: “AI is being applied in how we offer interactions with the market through Dynamic M-ELO and how we manage options products through strike optimization.” For the future, he cited potential AI opportunities in three areas: low-liquidity markets, such as exchange-traded product issuance; non fungible assets or having similar assets trade together; and regulatory technology where generative AI can improve compliance posture and reduce friction.
Stanford’s Pelger discussed AI and big data applications for investment and asset management, with a focus on equities.
Pelger said that machine learning can deal with high volumes of data and can capture complex relationships in a robust way, and at present, there is significant hype about how machine learning can improve investing. But it remains difficult to accomplish in the real world.
Pelger presented findings from four of his research papers, pertaining to statistical asset pricing factors, deep learning factor models, asset pricing trees and deep learning for statistical arbitrage. Overall, he stressed the importance of industry knowledge as a complement to AI applications.
“Machine learning can generate large investment gains if used right,” Pelger said. “To do so, you need to include domain-specific knowledge.”
“There is huge potential for using machine learning in investment and risk management,” Pelger concluded. “But you need to combine economic structure with machine learning… There can be limited success for off-the-shelf machine learning without financial understanding.”
Watch the Nasdaq Market Makers webinar replay here: https://www.nasdaq.com/market-makers#mmwebinar
