Quant Guide 2019: Cornell University

School of Operations Research and Information Engineering, Ithaca and New York City, New York, US

Master in Financial Engineering | metrics table at end of article

Ranked 10th in the Quant Finance Master’s Guide 2019

Cornell’s financial engineering master’s programme gives its recruits experience of learning in two contrasting environments. During the first two semesters, students are based in the leafy surroundings of Ithaca, NY, learning the fundamentals of quantitative finance. After a summer internship, they relocate to the hurly-burly of Manhattan for a final semester involving industry projects and networking.

The course structure is geared towards helping students prepare for the shift to the workplace, says programme director Victoria Averbukh. “When grads start their job, they have to be useful on day one,” she says.

The industry focus is evident before classes even start. During August and September new recruits take part in financial engineering “career bootcamps” in New York City, involving meetings with alumni and other practitioners. Another pre-programme element is the three-day MEng Connect, a course of intensive study designed by an industry expert.

“Our programme is designed as a professional degree,” says Averbukh. “From the outset, students are exposed to professionals in the financial industry, whether it’s through meetings, networking sessions, meet-and-greets or through practitioner-led curricula.”

The programme is offered within Cornell University’s School of Operations Research and Information Engineering. The first two semesters focus on foundational mathematics and computing. The curriculum for the third semester consists of seven courses, five of which are taught by industry figures who, Averbukh says, simply “teach what they do”. One course, entitled ‘From Models to Markets’, which covers the use of models in constructing trading strategies and assessing risk, is taught during evenings because of the instructor’s day job.

Taking the practical elements further, some course components are assessed by non-academic industry personnel.

“We have classes on asset management where students present work to a team of judges from the asset management space, who give them grades for the class,” says Averbukh.

Two years ago the programme introduced an optional certificate in financial data science for students interested in focusing on “big data and its applications, whether within risk management or alpha-seeking positions or just in research”. Over half of the latest cohort signed up for the certificate. Another recent development to the MFE was a “non-credit-bearing, completely optional web scraping workshop”. As banks and quantitative investment firms look to exploit new sources of data, students see a merit in such topics. “Out of 59 students, 47 showed up,” says Averbukh. “And then the next week they all showed up again for the more advanced version.”

After the first year, every student on the programme undertakes one of 10 industry-sponsored projects.

“They get this experience of pseudo-working,” Averbukh explains. “Many of them do the project in the offices of their sponsors. We make sure we offer a broad range of topics: natural language processing, risk management, commodities, fixed income, equities. And it’s not just mindless data work – there’s value.”

“I was looking for a curriculum which really combined theory and practice,” says Nicolo Perugini, who graduated from the programme in 2016 and now works in electronic trading at UBS.

“It felt like great preparation for the professional arena,” he adds. “Being able to work on a real-world problem during my project, collaborating with senior representatives from [the sponsor company] together with faculty advisers – it proved extremely valuable.”

A sign of the value that sponsors place on the students’ work is in the non-disclosure agreements that are required for many of the projects. For example, the natural language processing project for the 2018 fall semester is related to text mining in Chinese, but Averbukh is unable to give further details: “I can only disclose so much because a lot of students and faculties sign NDAs.”

The school has to weigh up the commercial realities of firms wanting to protect their intellectual property, against the need for all students and teachers to receive wider educational benefits from the projects.

“I’ve had to drop sponsors because their NDAs were too restricting,” she says. “Every project has a student team, a sponsor and a faculty adviser. And some sponsors were saying that outside of the team and the one adviser, nobody else could see the final report. I said, ‘That’s not going to work for me. I would like the Cornell community to have access.’”

Averbukh identifies AI proliferation and, more broadly, the progress of new tech as the forces driving change in the financial industries.

“Not many financial companies understand what the implications are of machine learning today. In the future, I see a more start-up, more fintech type of Wall Street.”

As the changes take hold, Averbukh urges academic institutions to keep pace by improving the way they collect and publish data, perhaps with a kind of trade repository of research: “I wish across departments, across schools, for a good, collaborative depository of papers and summaries of research by faculties so that master’s and PhD students could see what the faculty is actually doing, instead of listening to pitches like mine.”

Click here for links to the other universities and an explanation of how to read the metrics tables