Quant Research VS Quant Trading
Can anyone explain the difference between the two roles at a typical hedge fund? What does each one do and what academic background do they require? Also, how do their compensations differ?
Gonna copy and paste an earlier comment I made on this subject.
I think it becomes much easier to understand if we drop the unnecessary name inflation (with “Quant” added to all the roles).
Fundamentally at prop shops (applies to quant hedge funds too, but hedge funds differ in that there are fewer traders and the role is typically execution focused, with the alpha coming from researchers), there are 3 main roles - researchers, traders, and devs:
1) (Quant) Researchers: The ‘quants.’ These are typically PhDs or exceptional undergrads/masters, though some shops are more open to undergrads and some aren’t. Interviews are a mix of stats, linear regression, coding, and probability puzzles. This role doesn’t really work on market hours like a trader and is charged with coming up with strategies (‘signals’) and working on creating and coding the models that traders will use to trade live. Much more coding but if you like doing research and posing hypotheses/using the scientific method, plus prefer a slower pace than trading, QR is the way. If you have a successful model, you can expect to do very well. Less job stability and security than being a dev, but more than being a trader (very roughly). QRs are typically the lead at quant hedge funds and certain prop shops (Tower, HRT, Jump).
2) (Quant) Traders: Typically undergrad or master’s students. Also strong quantitatively, but not requiring the research background and rigor that QR requires. Typically need to be better at quick math and probability games interview wise, not much coding tested. Similar compensation wise to a researcher with a lot of variance (you will get compensated more if you have a stronger delta to results and this is more likely to happen at a trader led shop) plus the role can be pretty broad in how research heavy or trading heavy it leans. Live trading isn’t actually manually entering bids and asks but more like adjusting volatility curves on models live and responding to specific market situations/dynamics as they come up. Also you typically do a good amount of data science work in python (using pandas etc) and, if you desire, can do mostly research/projects depending on the firm you’re at. At Jane Street for example, traders do all desk specific research. Typically this role is the lead at prop shops (some that stand out as being pretty trader focused include Jane Street, Optiver, and SIG).
3) Devs (and Quant Devs): Some firms choose to delineate between developers and quant developers and some do not. Typically this role is gonna be standard software engineering interview wise, but maybe with some math/probability stuff thrown in. Devs at prop shops and hedge funds can expect to work on low latency C++ stuff, or make tools for traders / quants. Quant devs on research teams can sometimes implement strategies that researchers are working on. Still get compensated quite well (this year, Jane Street and HRT offered 375 year one to their devs, and Cit offered 380 as a return offer, though 150 of it is a signing bonus) and have much more stability to boot. Of course, this comes with less potential upside down the line.
Some of the points I made in describing the roles are more applicable to prop firms. As stated earlier, at many quant hedge funds traders are execution focused and unlikely to be contributing alpha, so they are not compensated as well as researchers. Prop shops are more equitable in compensation since the trader role has much more autonomy and ability to generate alpha.
Thanks a lot for your insight! So, the researcher basically comes up with a trading strategy (algorithm?) and the trader executes it during market hours (ie. tweak the variables)?
In a very reductive sense, yes. Strategies can be systematic or semi-systematic depending on the products being traded and a trader's role is going to vary based on the type of strategy being used. The researcher will use some set of data to find signals, run backtests, and if a strategy is put into production, then a trader could be simply executing and monitoring it (if it's particularly systematic) or might be making active decisions to adjust certain strategy parameters and hedge the book against certain events / etc. This isn't changing variables in code (not sure what you meant by that)—usually, traders have some proprietary GUI to move volatility curves or whatever. They will also do data analysis on certain trades and conditions using python. Also, at some firms researchers will write production-ready code, but at others, they will write (not production-ready) code and logic that quant devs will then implement properly.
There is no hard defined difference. Large quant funds typically divide responsibility between researchers, developers, and traders, but the roles can be quite fluid. At small fims, the roles will often be combined because there is not enough PnL to justify hiring so many bodies and building out so much infrastructure.
Research: Work on converting some proprietary raw data to a tradeable signal. Running backtests on that signal to understand how to best blend it into a portfolio of existing signals. Typically paid on a 50/50 base/bonus structure. Junior researchers don't have much PnL linkage, more senior researchers might get paid depending on their signal's performance.
Developers: Take the signal logic that research discovered and code it into the firm's software framework to robustly and seamlessly trade day-to-day. Developers will work on problems like: can we calculate the signal fast enough, what happens if data is missing, how can we get the signal to update reliably in real time? Typically bonuses are treated as deferred bases (not tied to PnL).
Traders: Observe the position recommendations in real time, question if they make sense, and make small tweaks to adjust for day-to-day market conditions. He actually is responsible for making sure the portfolio trades to the recommended position. Typically paid in a low base/high bonus structure. The bonus might be tied to active alpha, total PnL, or tracking error depending on the strategy.
Agree that these three roles get increasingly fluid as a firm decreases in size. Want to make a comment on devs never getting profit-sharing deals though.
Jane Street and Optiver do global profit-sharing, so devs are tied to PnL in the same way traders and researchers are. At Jump, embedded desk devs can get nice profit-sharing deals too. That said, I'm sure this never happens for hedge fund devs, only ones at prop shops. Also, at prop firms researchers usually have compensation leaning heavier towards bonus than to base, haven't seen enough data points for hedge funds but 50/50 base/bonus surprises me.
This is correct. Other descriptions of quant traders are more for prop shops. At most funds, quant traders role is in execution (minimize costs, maximize fills) and monitoring for issues in orders and extraordinary market conditions.
Tempora nam sapiente est et. Cumque quidem dicta corporis est cumque non perferendis. Eaque incidunt ut molestias dolore dolorum recusandae quaerat.
Eius amet repellat doloribus perspiciatis facere sed. Unde est eveniet qui eum. Vitae reprehenderit tenetur a ut pariatur tenetur.
Saepe quia dicta est. Sint earum voluptatem velit minima. Maxime dolorum totam laboriosam natus sed architecto eius ut. Aut voluptatibus quibusdam quas voluptatem maiores. Tempora voluptas quo qui accusamus magni tempore. Reiciendis sed ipsam facilis quaerat odio esse optio. Quasi optio temporibus consectetur ab fugit non.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...