What is a trading analyst position like?

[The Spartan Guy]

About myself:
Finishing Ph.D. in computer engineering from a good (but not top) university. Interested in quant positions, specifically those which involve lots of mathematics, stochastic calculus etc.


Initially, I was giving serious thoughts only to investment banks like J.P.M, Deutsche etc. But I found a few trading analyst positions from a head-hunting firm in some hedge funds/prop trading firms. Not being from the industry, I cannot really judge the positions from their little details. The details look like*

XXX company actively make markets, as Designated Market Makers, across a broad range of asset classes including equities, foreign exchange, commodities, options and fixed income, providing two-sided liquidity on over two hundred market centers around the world.

In another company, YYY, all their trading is highly structured and based on an algorithmic approach executed in automated manner. YYY utilizies state-of-the-art proprietary software and advanced valuation models that allows it to be one of the most competitive market-makers in the major derivatives markets in Asia. With a focus on discipline and risk management, YYY is able to profit from all types of market conditions. Their strategies exploit inefficiencies in the market and utilize a multitude of high frequency trading techniques that have low risk and provide high-return.


What drew me to the job market in the industry are a few career talks by investment banks (which were incidentally given by Physics, Math Ph.D.s), but now I see most of the openings are like this. So I would appreciate some guidance on the comparison between these to apparently different (but may be same) job profiles. In particular,

1. How does a quant's job differ at a place like J. P. Morgan to one of these trading firms? Typically at sell side banks I see the job titles as quantitative analyst, associate, risk analyst etc. But in the buy side firms, the titles are like junior traders, trade analyst etc. Are they just euphemisms?
2. As a computer engineering Ph.D. (with some knack for mathematics, stochastic calculus, some economics etc.) which one do you think will be more interesting to me? I know it depends on many factors, which can change with time. But be as general or as specific in your answer as you feel given this information.

Some guidance from the industry veterans and many experienced people present here will be really appreciated. I asked the guy from the head-hunting firm, but it seems his own knowledge in the specific jobs that quants do is quite limited.


*Perhaps you can already guess the firms' names from the profiles and descriptions. But still I did not want to name them directly.

 

[prooney]

Typically the quant at a bank is building valuation models while a quant at a prop shop is more actively involved in the trading activity. Sometimes stationed at the trading desk itself. The prop job is likely more stressful as your dealing hands on with traders who may be stressed and irritable however the payout is likely greater at the prop shop.

 

[financeguy]

It sounds like the jobs you are looking at in banks are for quants supporting trading desks, while the jobs on the buy side are quantitative trading / market-facing jobs (whether automated market making or stat arb or whatever). Normally I would advise people looking at trading jobs straight out of school to go sell-side first, learn the market properly, then lateral over to the buy-side to take advantage of getting a formulaic payout on a pnl, which the sell-side no longer provides. Sometimes if you join the buy-side at the entry level, you don't get the sort of support you may need early on that a large investment bank has institutionalized, which in turn makes it more difficult to move up the ranks organically and claw a bit of capital for yourself to manage. However, it sounds as if you are not comparing like-for-like. Being a quant supporting a trading desk at an investment bank is not market-facing, and it is not usually straightforward to transition from that kind of seat into a buy-side trading seat. Just be thoughtful in differentiating the buy-side firms in terms of culture of supporting junior traders' careers and timeline to getting promoted (hopefully they have a transparent track record of doing so).

 

[Gavin L]

Sorry if this is off-topic, but responding to financeguy, what would you think is an appropriate time to spend at a sell-side bank before going buy-side? Is the traditional 2 year stint for IBDs going to apply for quants?

 

[financeguy]

Sorry if this is off-topic, but responding to financeguy, what would you think is an appropriate time to spend at a sell-side bank before going buy-side? Is the traditional 2 year stint for IBDs going to apply for quants?

There isn't a traditional timeline or route like there is for private equity. As I understand it, those would normally spend 2-4 years in an IBD on the sell-side (essentially completing their analyst or associate program plus or minus a year), and then try to make a move to a private equity or hedge fund, doing similar sorts of fundamental corporate analysis for more upside. On the trading side, like from a fixed income or equity trading floor, there isn't a standard path like that, primarily because historically the sell-side would pay and retain its top risk takers in these areas rather than lose them to the buy-side. Obviously now that isn't the case for a variety of reasons and people are moving en masse to the buy-side. Traders would move to the buy-side, but it wasn't systematic, and juniors were hired as needed.

The point I'm trying to make is that because of how the market has worked historically, there isn't a system in place whereby hedge funds recruit sell-siders with 2-4 years of experience and have training programs and a willingness to develop the skills of promising young traders. That is now changing as the talent pool from the sell-side is continuing to dry up, and hedge funds can't rely on poaching from investment banks as a business model for hiring practices. But while we are stuck in this in-between phase waiting for firms to adapt to the new world order, what I'm getting at is that it is a subjective decision. I can give you my opinion but there's no right answer. I think the right amount of time is 3-5 years. The first 2 years of a trading job, you're still figuring the basic things out - everything from the details of your job and the micro details of your market to how it fits into the overall business your group is a part of. I think if someone leaves in under 2 years, they are leaving only partially trained. After 2 years, if you spend at least 1 more year, then you have a year under your belt of being a real contributor (hopefully) to the business. A couple more years of doing that, great. After 5 years, I think you start running into the issue of seniority. A hedge fund will hire people to support the portfolio management process (e.g. execution traders, idea generators, etc.) or to actually run a portfolio (i.e. portfolio managers). To be the former, anything 5 years and in works. To be the latter, coming from the sell-side, you need more like 8-10 years of experience. The point being that after 5 years, you enter a period during which you're too senior for junior-mid level roles at hedge funds, but too junior for senior roles. So that's how I get to the 3-5 year window.

The above is for traders (quantitative or otherwise) moving to a hedge fund for a similar sort of role in risk taking. If we're talking about quants on the sell-side becoming quants on the buy-side, I would say probably more like 5-7 years is a sweet spot to call yourself an expert on something, but not so senior that you're above doing the implementation work yourself. But a big caveat here is that I am not a quant, I am speaking here as someone who works with quants and knows a lot of them, but haven't actually gone down that path myself.

 

[Gavin L]

There isn't a traditional timeline or route like there is for private equity. As I understand it, those would normally spend 2-4 years in an IBD on the sell-side (essentially completing their analyst or associate program plus or minus a year), and then try to make a move to a private equity or hedge fund, doing similar sorts of fundamental corporate analysis for more upside. On the trading side, like from a fixed income or equity trading floor, there isn't a standard path like that, primarily because historically the sell-side would pay and retain its top risk takers in these areas rather than lose them to the buy-side. Obviously now that isn't the case for a variety of reasons and people are moving en masse to the buy-side. Traders would move to the buy-side, but it wasn't systematic, and juniors were hired as needed.

The point I'm trying to make is that because of how the market has worked historically, there isn't a system in place whereby hedge funds recruit sell-siders with 2-4 years of experience and have training programs and a willingness to develop the skills of promising young traders. That is now changing as the talent pool from the sell-side is continuing to dry up, and hedge funds can't rely on poaching from investment banks as a business model for hiring practices. But while we are stuck in this in-between phase waiting for firms to adapt to the new world order, what I'm getting at is that it is a subjective decision. I can give you my opinion but there's no right answer. I think the right amount of time is 3-5 years. The first 2 years of a trading job, you're still figuring the basic things out - everything from the details of your job and the micro details of your market to how it fits into the overall business your group is a part of. I think if someone leaves in under 2 years, they are leaving only partially trained. After 2 years, if you spend at least 1 more year, then you have a year under your belt of being a real contributor (hopefully) to the business. A couple more years of doing that, great. After 5 years, I think you start running into the issue of seniority. A hedge fund will hire people to support the portfolio management process (e.g. execution traders, idea generators, etc.) or to actually run a portfolio (i.e. portfolio managers). To be the former, anything 5 years and in works. To be the latter, coming from the sell-side, you need more like 8-10 years of experience. The point being that after 5 years, you enter a period during which you're too senior for junior-mid level roles at hedge funds, but too junior for senior roles. So that's how I get to the 3-5 year window.

The above is for traders (quantitative or otherwise) moving to a hedge fund for a similar sort of role in risk taking. If we're talking about quants on the sell-side becoming quants on the buy-side, I would say probably more like 5-7 years is a sweet spot to call yourself an expert on something, but not so senior that you're above doing the implementation work yourself. But a big caveat here is that I am not a quant, I am speaking here as someone who works with quants and knows a lot of them, but haven't actually gone down that path myself.

That was so incredibly informative... thank you - I will continue to read this as I start my summer internship at a BB sell-side bank.

 

[Yike Lu]

Speaking strictly to the point of @The Spartan Guy being a PhD in Computer Engineering, it's much more likely that a buy side firm will have better IT environment for you. I've worked on both sides of the fence, and the one time I was at a sell-side shop, the IT situation was massively restrictive and painful.

 

[The Spartan Guy]

Speaking strictly to the point of @The Spartan Guy being a PhD in Computer Engineering, it's much more likely that a buy side firm will have better IT environment for you. I've worked on both sides of the fence, and the one time I was at a sell-side shop, the IT situation was massively restrictive and painful.

Thanks so much. On a related note, even though I am in the computer engineering department, rather surprisingly, I am interested more in the mathematical side, like stochastic calculus for example. Do you think there can be some opportunity to engage in that if I am in the industry? Or is it entirely about coding?

 

[Yike Lu]

The blue skies pure math research is ever more rare in industry. The areas where that was applicable (exotics, mortgages, credit) are declining due to bank regulation. To be fair, they may arise again in a different form, so long as people want to trade these products.

I do a fair bit of mathematical work, but it's more in the areas of optimization, basic financial math (Black Scholes extensions), and data science. These days coding is necessary both to conduct the research and implement the result. It's highly unlikely that you'll be getting paid writing a paper on a new stochastic volatility model -- too far from the bottom line and also you're the wrong PhD for it.

 

[pingu]

Thank You for Post, I am interested more in the mathematical side, like stochastic calculus for example.

sorry to hear that.