From PhD to Quantitative Researcher at Optiver

Q: You earned a PhD at the University of Cologne before joining Optiver. What did you study?

The field of physics I focused on is called statistical physics, which deals with properties of large groups of objects, mostly related to solid state or gases. In that specific area, I focused on random matrix theory. It’s a mathematical framework used to describe systems where there is a lot of randomness, a lot of noise and chaotic behaviour. You’re trying to extract information despite all of this chaos and uncertainty.

Q: How does random matrix theory apply to financial markets?

The math behind random matrix theory is quite universal and can be applied to many things. Portfolio optimisation problems in financial markets, for instance, where we observe many timeseries of stock prices, each seemingly random. As soon as you have many sources of signals, you get a lot of noise from each of them. The challenge is trying to extract as much useful information as possible from multiple streams of data. Q: How much did you know about finance before coming to Optiver?I had a good understanding of the math and models, but I had no idea about the subject of finance in and of itself. But this is the part that we’re happy to teach people on the job, and it’s easy to get up to speed.

Q: What’s a typical day for you?

I lead the pricing research team, which builds the libraries we use to transform the inputs we get from trading teams into the actual prices of derivatives that we show to the world. There’s no typical day, every one has its own challenges. Some days you’ll spend just reading academic literature, exploring a particular topic. Another day could be spent developing an optimised algorithm for a specific partial differential equation solver. The major challenge of our work is that we really need to verify whether something works in the real world – that’s one difference from academia. So we’re going to be testing our solutions on real world data, assessing whether they will actually work in 99.99% of scenarios. That’s part of the process of rolling out new algorithms and pricing engines: gathering feedback from traders, gathering metrics for how they perform, trying to fix bugs as they occur.

Q:  What is a misperception you think people have about quantitative research roles?

Aside from the hard skills – programming, statistics, probability, numerical analysis – academic literacy is extremely important in a quantitative research role. Being able to comprehend a complex problem, dissect it into parts, understand it inside out – that is something that you really learn while doing an advanced degree. When mistakes are very costly, you want to have people who are reliable and have the ability to assess whether a solution as a whole is going to work.

Q: What is a misperception you think people have about quantitative research roles?

One thing I would emphasise is that this is a really collaborative environment. I used to think this was a cutthroat industry where everybody was out for themselves. But people are looking out for each other, thinking about the best way to help others.One other misperception that people have about quants in prop trading firms is that we merely get tasks from trading and execute on them. That’s very much not the case. This is a collaboration with trading, not a top-down or hierarchical set-up. It’s very much about getting feedback, improving, communicating and educating each other in the process.It’s definitely intellectually stimulating, too. It’s not too far from academia in that way. The problem space is slightly different, but you’re going to scratch the same itch at Optiver.Interested in a role as a Quantitative Researcher at Optiver?