Stevens Institute of Technology

01/10/2025 | News release | Distributed by Public on 01/10/2025 10:11

Stevens High Frequency Trading Competition Serves as Springboard for Research and Learning Opportunities

Student Spotlight

Stevens High Frequency Trading Competition Serves as Springboard for Research and Learning Opportunities

posted

Two Stevens graduate students and former Stevens High Frequency Trading Competiton play a key role in hosting the annual event

Transactions are not the only thing the Stevens School of Business annual High-Frequency Trading Competition produces in large volume. In recent years, it has inspired students to seek advanced degrees and work on research within the financial engineering and financial analytics and technology domains.

Two competition alums who are pursuing a master's and Ph.D. at Stevens are assisting with the 2025 competition. Adam Moszczysnki completed his bachelor's degree in quantitative finance last spring and is pursuing a master's degree in financial technology and analytics. [IR1] Matthew Thomas also graduated from Stevens in 2024 with a bachelor's degree in computer science. He is currently working on a Ph.D. in financial engineering.

Both credit their exposure to this type of trading and the SHIFT high-frequency simulation platform created by Dr. Ionut Florescu, program director of the financial technology and analytics program at Stevens, for their current path.

"I was a first-year student at the time, so I wasn't very versed in market microstructure or anything like that," Adam explained. "I only took some basic level classes, so it was a really good introduction to trading itself and what it takes to make a profitable strategy in a realistic situation. It was an eye-opening experience, and it took a lot of work to get our bearings and actually create something that worked. It's incredibly interesting, and I like the master's program specifically because it gives you a more in-depth understanding of what's going on with higher-level statistics and the math behind the markets."

But a master's degree is just the starting point.

"The competition solidified my interest in these things, specifically high-frequency markets," he said. "Coming out of the competition, the research with Professor Florescu was about all the things I was eager to get into and learn more about. It put me on a path of learning more and getting more experience with these markets, ultimately leading to that desire to get a Ph.D. There is always more to learn, especially with very advanced statistics. You can always create a better model or better understand the markets as a whole, which is something I'm interested in and why I got involved in the competition from the start."

While not enrolled in the School of Business, Matthew began his trading journey with the Stevens Student Managed Investment Fund (SMIF). He and his team won the competition in 2022.

"It was a really cool experience for me because coming from computer science I was fairly unfamiliar with most of how financial markets work, especially at the high-frequency level," Matthew said. "Doing this competition was a great learning experience for me. There was a lot to learn and many difficulties and challenges to overcome, but we ended up winning the competition. That led to research opportunities and, ultimately, to the path I'm on right now. I study the application of reinforcement learning to simulate financial markets to understand how they work, specifically for high frequency."

"Reinforcement learning is a type of machine learning in which an agent can interact with some sort of environment, get feedback from the environment, and adjust its strategy accordingly," he continued. "It's used to solve a lot of games like chess, but it's also able to solve our big financial game, which is trying to profit from financial markets. We model agents that submit high-frequency orders, and we analyze the markets that emerge. In doing so, we're able to make realistic simulations."

Adam's experience as a participant and organizing team member will be a valuable resource to this year's competitors. He has worked on the underlying code that forms the basis for the event and will serve as a "lifeline" for those who need help getting started and setting up their Python application programming interface (API).

"I'm specifically helping everybody get started setting up their versions of the machine so they can test their strategies before actually putting it into production," he said.

After his team's victory, Matthew helped with overhauling the scoring system to its current "Formula 1" format. The competition occurs in five stages, with increasing points awarded for each portion. During each stage, teams have a set of parameters to create algorithms that they'll put to the test on the SHIFT platform each Friday. Points will be awarded based on how each team finishes compared to the other participants.

"We started this style of scoring system last year," he said. "Teams get points based on their ranking on each competition day, and the number of points increases for every subsequent competition day. That's different from how we did things when I was in the competition. It was purely based on how much profit you made or how much money you lost. That meant you could lose a little bit of money every day, but as long as you made enough to make up for it by the end, you could win."

"That's how we won, but we felt that maybe it wasn't as fair as it could be," he continued. "We want students to value every single competition day. The ranking on each day matters, but we also want students to be able to come back from doing poorly at the start of the competition. This scoring system covers both of those principles for us, and I think it's very fun. It's a really engaging way to go about this competition."

Applications must be received by February 1, 2025, and teams will be notified of their status by February 15. If selected, there is a non-refundable registration fee of $100 per team. A mandatory information session is scheduled for March 3, and teams will be able to test the technology from March 7-10 before competition trading begins on March 24. Trading concludes on April 25.

Applicants must be domestic undergraduate or graduate students currently enrolled in any university located within the U.S. They can be pursuing an undergraduate or graduate degree in any subject, but some members of the team must be familiar with Python programming and Github. Each team must be between 2-4 members.

The winner of the competition receives a $2,000 prize. The second-place team receives $1,200, and the third-place team earns $400.

For additional information, please contact [email protected].

[IR1]In the previous sentence it says they are both current Ph.D. students but here it says he is pursuing his master's degree. Just need to clarify or correct.