Hello,
This is a hypothetical at this point, but if one were to develop a trading strategy based upon research done in graduate school (Master's level), would he or she be able to trade this strategy at a proprietary trading shop?
I watched a presentation from Tom Hutchinson from Belvedere Trading and had the sense that new traders are trained in how to trade Belvedere's strategy, rather than tasked with developing their own. Is this understanding correct? And if so, is this common for the junior/intern level quantitative trading roles for prop shops?
Any insight would be greatly appreciated. Thanks.
It’s quite unlikely they’ll take you seriously without a multi year track record in a professional setting. Trading your own account or paper trading usually doesn’t hold much weight. My thought process would be to do great research and market yourself to firms that are interested in strong researchers.
Okay that's interesting. My sense was that traders would have more opportunity to implement strategy because they are responsible for their own p&l.
But it sounds like that may not be the case. Is the idea that junior traders are kept on a short leash but researchers are tasked with developing strategies?
I think the first thing to understand is you won’t be solely responsible for PNL without 3-7 years of experience at the firm or prior professional experience being responsible for PNL.
So to transition into a role where you could be solely responsible for strategy generation or PNL you have to start as a junior trader under a trader (shorter path) or researcher/ operations (longer path).
The point I’m trying to get across is don’t think you’re going to walk into any firm and start building/implementing strategies at best you can initially assist with the strategies they’re already building. The attitude of I want to build out my ideas will come across as arrogant, even if your ideas are a gift from god.
So if you’re doing a masters in a quantitive discipline get involved in groups that have connections to prop firms, write a paper on your investment research, show an interest in markets, and show a commitment to the companies goals/ direction.
To directly address your comments traders and researchers work together on new ideas. Who drives new strategy development varies on firm and individuals, so no good answer.
I think track record is always a plus but firms like Geneva Trading (Chicago, Dublin) are seemingly always open to new strats. Very much an eat what you kill vibe (less c. $2-3k in monthly desk costs), but if you have the conviction is certainly an avenue to explore.
Can you elaborate on this? At Geneva one could opt to pay $2-3k in monthly desk fees for the ability to trade using Geneva's infrastructure? Would one have to provide their own capital?
I'm very much looking for an eat what you kill vibe.
there are different prop shops with varying payout structures...depending on what you trade.
are you trading equities and require the ability to borrow shares? are you trading interest rates? could you trade just futures?
depending on the answer will determine the best infrastructure.
if you actually have a strategy...and its is both profitable, and has very small drawdowns, backtested over the last 10 years...then sure, you could trade that at a prop shop. You would start off very small, and then gradually increase size as profits accrue. You could get a job as a quant researcher, and then once you are inside a firm, you could talk to the traders, show them your strategy results, and ask for a small account to trade the strat and see how it does. I've seen this happen, and i've seen if fail.
However, i think you underestimate the difficulty in coming up with a new strategy that is consistently profitable, that also has minimal drawdowns.
I haven't estimated it at all, at least not in the OP.
This ... sounds ... so ... scammy. I know you watched Limitless and all that, but why why why would anyone do this for real? As per my understanding, trading strategies are of two kinds.
The Upside Down Blind-Folded Triple Lutz Oscillator might sound good on paper but there's probably a reason that you came up with it in grad school, just like everybody else does - it s8x0r.
As our esteemed colleague, Monsieur faceslappingcompilation has noted, there are some very basic tests to check if an algorithm is non-retarded at even a cursory level. K-fold (step forward) cross-validation is but an initial test to get you started.
What kinds of algorithms and tests do you think are being used by the people on the other side of your trades?
Why not just work at [any other job] and use your savings to try out this strategy on your own with a few bucks to start?
Any decent prop shop will let you trade your own strategies if you've proven them to work. The amount of "dues paying" you have to put in before anyone will listen to you will scale with the size of the firm.
At a place like Chicago Trading Company it may take you 3 years before you're allowed to put on your own ideas. At Trillium, you're expected to have your own strategies after 3 months.
Yes, but if you're a grad student, chances are that this strategy is already being traded there or that this strategy has been tested by someone else but doesn't execute IRL as well as you think. You're a smart guy to get hired as a QR at a Chicago prop shop, but 95% of campus hires aren't coming in with stuff that is as novel or unique as they think.
There is definitely room for creative thinking and inspiration in equity systematic strategies, but in another sense there is also very little that is new under the sun. If you're thrown into a systematic portfolio with a number of signals, the strategies that most Master's and even many PhD students-- short of a JoF publication-- would come up with have probably already been implemented.
Ok, so the way to get ahead as a campus hire isn't creativity. At least not yet. So how do you become more valuable as an entry-level systematic quant?
1.) Understand the different market participants, and what they are doing. Get a good idea of the big picture. 2.) Get a good sense of what sorts of data makes money, and how it makes that money. 3.) Understand your stats, and how the different analytics of signal performance interact. Also, if you try 20 different different signal ideas on random noise, what are the odds that you get a t-stat of 2 on one of them? 4.) This isn't the sell-side. Alpha Devs on low frequency signals executed through sell-side brokers have to be right 52% of the time, not 99.99%. Mistakes generally aren't as disastrous here-- the key is iterating fast, trying to deliver what was asked of you (along with another related insight or two if appropriate) and keeping the back and forth with the PM or Senior Researcher going. We're generally not making pitch decks for banking clients, nor are we executing 7000 contract orders where a penny costs thousands of dollars. 5.) Talk to people on other teams. Have some general interest in what they're doing. A lot of the time there's ways to get different teams to work together on stuff, and making one big pie bigger than two smaller ones. 6.) Always give attribution for ideas. Like with any citation, it rarely hurts you, it almost always helps them, and makes the overall pie bigger. It makes you the kind of person that people want to work with and share their ideas with.
There are exceptions, but generally, the rule is that nice people are the ones who succeed at the rank-and-file level in quant research. (It gets a lot more complicated when it comes to management, but we'll get there when we get there.)
At the intern level, they might have you backtest some strategies, but the thrust of your internship will probably be your boss's idea. It will probably also be productionized by a FT employee, possibly you when you get back (just due to the nature of the FT support work involved.)
Thank you. This is an insightful response. Going to send a PM later today.
Sed optio ut quam corrupti aut tenetur. Reprehenderit aut quasi explicabo sint voluptatem. Laborum unde dolorum eius rerum tempore blanditiis dolorem.
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...