Forums  > Careers  > Switching from Data Science to Quant Research?  
     
Page 1 of 1
Display using:  

M. Friedchickenman


Total Posts: 8
Joined: Jul 2020
 
Posted: 2021-03-23 19:40
Hi everyone.
Currently a DS at a major financial company.
Compensation is good and I work on interesting and unique problems dealing with capital markets. (I already work with lots of market data and solve problems that could be used to make investment decisions if the company wanted to? But conflicts of interests. In a way, I build products and services that could help make better investment decisions.)

Recently I became disillusioned with the whole DS role at a commercial enterprise. Research only needs to be "good enough" and there is more of a focus on delivering a product than getting it right.

Obviously, the objectives at my organization is not to make good investment decisions but to build products and services that are marketable.

I've been thinking about switching to Quant Research so I can be just another highly paid research person and not worry about "business goals" and presentations. I want to focus on doing really good research, not satisfactory research attached to product delivery.

My questions are
1) Would I even get interviews? (I only have a BS)
2) Is buy side Quant Research really a pure research role or is there more to it?
3) Is the work actually interesting?
4) Am I kidding myself?


marcmiak


Total Posts: 1
Joined: Mar 2021
 
Posted: 2021-03-24 00:44
I am in the same situation as you and has done some research on it.
Just to give more background, I hold a Ph.D. in Statistics and have been working as a data scientist for seven years in the insurance industry.

I am looking to switch from data science to quant research roles on the buy-side.
My plan is take the EPAT certification by Quantinsti and the Quantnet C++ certification.

I am curious to see what folks walking in the industry think about this plan.

Are you planning on taking any trading or quantitative finance certificate?


M. Friedchickenman


Total Posts: 8
Joined: Jul 2020
 
Posted: 2021-03-24 02:05
EPAT? Never heard of it.

From what I can tell, QuantNet is a useless forum nowadays and there really is no need for certificates.

I have friends in the Quant buy side space (mostly as traders) and absolutely none of them thinks you need some "Certificate".

Might depend on the fund type though.

Its Grisha


Total Posts: 92
Joined: Nov 2019
 
Posted: 2021-03-24 07:14
I respect your motivations, I too have been very annoyed by sales driven research in the past. Work that results in trades is a different level of intellectual honesty in my opinion.

> Would I even get interviews? (I only have a BS)

You are probably looking for roles more along the lines of "quantitative trader" at quant trading firms than "quantitative researcher". In my experience for the researcher roles they are really looking to hire people with graduate degrees.

I've got a buddy with only a BS who holds the former title and enjoys the job very much. He is more on the side of being responsible for PNL, crunching orderbook data, and figuring out the best quoting game to play. But the actual pricing models are out of his domain and built by the PhDs.

Personally I don't think most trading strategies require a grad degree to implement, but that is the state of hiring when there is a long list of candidates who have one.

I might add there are lots of data science roles with asset managers and longer horizon hedge funds looking to use more alternative data sources. While it may not be the quant trading thing you are picturing, it is easier to get into and the purpose is still "to make good investment decisions" as you say.

> Is buy side Quant Research really a pure research role or is there more to it?
> Is the work actually interesting?

All about being at the right place, you might be a glorified sql query debugger in one shop or have full autonomy to build cool models in another. Too hard to generalize, I would pose lots of questions to any team you are interviewing with to figure out what the junior roles typically entail. It might mean taking a job at a smaller firm to have more interesting work.

> Am I kidding myself?
If you have a list of 3 top firms you might be. If you are willing to be flexible with respect to brand name I'm sure you can make this transition.

"Nothing is more dangerous to the adventurous spirit within a man than a secure future."

M. Friedchickenman


Total Posts: 8
Joined: Jul 2020
 
Posted: 2021-03-24 15:09
Thanks for the response. But I'm afraid that quant trading is really not for me. I did in fact interview with couple shops before (Jane Street, SIG, IMC, and the likes) during my undergrad and didn't like the whole "trader" lifestyle. (Interview process was fun though). It all seems to be thinking on your feet all the time and no sitting back to think deeply.

I wasn't really focused on TC then so I picked the closest thing to research, which was DS.

From what I can tell there are good number of firms that do hire quant researchers without a graduate degree (exceptional bachelor's or experienced hires). Namely Citadel does so and there are handful number of smaller ones.

The last thing I want to do is join a poorly performing fund or one with no learning opportunity. But I'm not sure where I can get this information.

naturalemojiprocessingalpharesearcher


Total Posts: 8
Joined: Mar 2021
 
Posted: 2021-03-25 05:30
I’m a Quantitative Analyst/Data Scientist at a longer-term manager which is still focused on making investments as Its Grisha has mentioned, and am trying to build up a valuable skill set that will allow me to make the same transition. But it’s hard when you’re the only quantitative person (though onboarding our intern last summer and dept looking to hire a senior in near future so I’m told) and you have to do the dance of office politics to convince fundamental investment managers of your ideas and to act on them. Mentorship hard to come by but I get that directed self-learning is the bulk of learning anyway as your manager is incentivized to keep you as source of labor that doesn’t know any more than you need to. That fundamental managers are easily impressed by my formulas and statistics because they are formulas and statistics is not a good sign. Definitely not trying to end up at a shitty fund or one with no learning opportunity or future either. But then again I feel like you could join a role in a brand name prestigious firm where that is still the case.

My only two no-nos are 1. bullshit (commercial DS has a lot of this as OP mentions, we have a lot in common w.r.t. attitude towards research) and 2. low performance. Given the absence of those two red flags, brand I could care less.

Maybe the truth of the matter is that in 2021 the vast majority of top performing firms are more engineering- and trading-driven than research-driven.

nikol


Total Posts: 1385
Joined: Jun 2005
 
Posted: 2021-03-25 15:47
> vast majority of top performing firms are more engineering- and trading-driven than research-driven.

Research is in Academia.

Top performing firms are about engineering and trading, true. These firms are open to new ideas they are eager to test. Not research in academic sense.

How else can it be ???

... What is a man
If his chief good and market of his time
Be but to sleep and feed? (c)

Jurassic


Total Posts: 415
Joined: Mar 2018
 
Posted: 2021-03-25 18:09
Whats the difference between a quantitative analyst/ds vs quantitative researcher? what assets are you in?

Maggette


Total Posts: 1322
Joined: Jun 2007
 
Posted: 2021-03-25 19:19
"...at a longer-term manager..."

"only two no-nos are .....and 2. low performance"

As a data-scientist you are probably aware of the fact that, depending on the definition of long term, this is incredible hard to classify ;).


Ich kam hierher und sah dich und deine Leute lächeln, und sagte mir: Maggette, scheiss auf den small talk, lass lieber deine Fäuste sprechen...

naturalemojiprocessingalpharesearcher


Total Posts: 8
Joined: Mar 2021
 
Posted: 2021-03-25 20:27
No difference just naming convention. I guess just QR more often implies work on trading strategies in liquid asset classes. And having a PhD.

I'm in private equity.

EspressoLover


Total Posts: 488
Joined: Jan 2015
 
Posted: 2021-03-26 02:12
At least in America, plenty of people without a graduate degree hold a QR title at "top firms". Nobody cares as long as you can make money. Don't get me wrong, a PhD can still help get your foot in the door because it's a credible signal of intelligence. But it's definitely not the only way.

As for QA vs QR, it's largely a distinction without a difference. As others have said, everything is engineering these days. I think in the 2000s this was different, because there was a Cambrian explosion of new products and strategies that kept requiring entirely new models to be built from scratch.

Nowadays, there isn't really anything new. We're pretty much refining the same shit we've been doing for the past decade and a half. There's no "pure research" left. Just tinkering and implementation. You're way more likely to spend your wrangling a Spark cluster than solving differential equations. To the extent that there are any pure research positions left, they're probably a low-impact career deadend. Adding one more Levy term to a jump diffusion model is not exactly moving the needle for the average desk at Goldman these days.

If you're really interested in doing academic research in an industry setting, I'd go for tech not finance. Somewhere like Facebook Research really does run like an academic department. A position there will let you do "pure research", and actually get rewarded for it. Nothing like that exists in finance anymore.

Good questions outrank easy answers. -Paul Samuelson

naturalemojiprocessingalpharesearcher


Total Posts: 8
Joined: Mar 2021
 
Posted: 2021-03-26 02:59
I do like this tinkering and implementation aspect. There are less expectations than for coming up with something completely from scratch, I don't have a PhD and am overestimating my ability if I said I could do that without fucking up many times prior, going to guess that most employers will not have patience for this.

Others on this forum have talked about the cult-like hive mind in tech and I am not interested, prop trading is probably more tech-driven than tech (many companies run by their marketing team). I'd rather make 90k a year doing supply chain logistics/OR kind of work than 250k as a "Research Engineer" at Facebook AI to deal with the most insufferable sychopantic (how else can you get into a top PhD in Machine Learning program these days?) people on the planet.

nikol


Total Posts: 1385
Joined: Jun 2005
 
Posted: 2021-03-26 08:57
It's viral

https://www.cnbc.com/amp/2021/03/22/goldman-ceo-addresses-junior-bankers-complaints-after-survey-goes-viral.html

... What is a man
If his chief good and market of his time
Be but to sleep and feed? (c)

prikolno


Total Posts: 94
Joined: Jul 2018
 
Posted: 2021-03-26 12:00
> I did in fact interview with couple shops before (Jane Street, SIG, IMC, and the likes) during my undergrad and didn't like the whole "trader" lifestyle.

DE Shaw, HRT, Citadel, TSI are some firms that come to mind in no particular order if you don't like the SIG, IMC, CTC, Optiver-type culture.

> 1) Would I even get interviews? (I only have a BS)

Depends more on what you've done. You can also migrate from a quant developer role to a research role. There are places that use titles such as "research engineer" or "quant researcher" to refer to a quant developer, so it's possible to repackage your resume through these.

I've used the same Greenhouse and Lever platforms that most firms use for screening and it's not like it automatically weeds you out because of your degree. I almost never pay attention to the degree, usually there's much more obvious signals. YMMV.

> The last thing I want to do is join a poorly performing fund or one with no learning opportunity.

I feel this is a mentality that junior job seekers get tragically wrong. A poorly performing place is actually a great place to be at if you're optimizing for learning opportunity and growth.

1. This means there's a high chance that your manager or senior quits, and you get elevated to a higher position because you're at the next index on the bus, with more leeway to fail and easier pay raises because they can't afford to lose you.

2. Unless you are exceptionally talented, one of the strongest predictors of future success and income in a trading career is the number of genuine connections you pick up across the industry. I emphasize genuine, meaning people who can share information under the table who are not headhunters or interviewees from other firms. And the window of opportunity for you to do so is quite small - I'd say a mean of 2 jobs. Once you get past junior PM level, it gets harder because people start reporting to you; the relationship dynamic becomes different, and so on. At a stable firm with no turnover, this route closes and your remaining out is to be exceptionally talented.

3. There's also the 2008 effect. After a good year is when firms sometimes make the mistake of bloating their headcount with lots of useless people to be your contemporaries, and when the top guys to learn from (usually your immediate manager who had a role in the earlier success) cash out or get poached, and leave the firm.

4. The least competent people are usually the first to jump a sinking ship. Sadly for them, this also creates a positive feedback loop because being around to see your things break and maintaining or scaling legacy crap that has hackily survived >3 years is one of the experiences with the highest market values. Joining a poorly performing firm is awesome because it often exposes all of that even to the new employees, and you're forced to figure out what actually makes money.

And after you've taken several arrows in back, notched enough cumulative gross losses that you no longer have a puke point, one day you wake up and it's like a switch has flipped. It's a certain je nais sais quoi - you become a little grumpy and jaded, but on the positive side, how well the firm is performing doesn't matter so much anymore, because whatever firm you join is just going to generate some PnL anyway.

M. Friedchickenman


Total Posts: 8
Joined: Jul 2020
 
Posted: 2021-03-26 14:22
> DE Shaw, HRT, Citadel, TSI are some firms that come to mind in no particular order if you don't like the SIG, IMC, CTC, Optiver-type culture.

Do you mean this as in firms like Citadel and DE shaw are more focused on problem solving skills but not so hung up on speed? At least that's my knowledge given what I've heard from friends at JS type.firms vs Citadel type firms.

> I almost never pay attention to the degree, usually there's much more obvious signals.

That's interesting. What would you say are key qualities you and other resume reviewers are looking at?

> A poorly performing place is actually a great place to be at if you're optimizing for learning opportunity and growth.

This is another very interesting thought as well. I've always thought that joining a place that's struggling or is chaotic is a great career and skill builder for proactive and entrepreneurial minded people (I categorize myself in this group). You can be part of the team that turns things around by taking bigger but calculated risks and creating structure from nothing. Really fun work as well. I essentially work at a place like this now (more on the chaotic just starting up end not struggling end) and it's been a great learning experience so far.

But then I also think that there's a difference btw a "sinking ship" and a "damaged ship" that can be repaired. Where would you draw the line?

M. Friedchickenman


Total Posts: 8
Joined: Jul 2020
 
Posted: 2021-03-26 14:22
> DE Shaw, HRT, Citadel, TSI are some firms that come to mind in no particular order if you don't like the SIG, IMC, CTC, Optiver-type culture.

Do you mean this as in firms like Citadel and DE shaw are more focused on problem solving skills but not so hung up on speed? At least that's my knowledge given what I've heard from friends at JS type.firms vs Citadel type firms.

> I almost never pay attention to the degree, usually there's much more obvious signals.

That's interesting. What would you say are key qualities you and other resume reviewers are looking at?

> A poorly performing place is actually a great place to be at if you're optimizing for learning opportunity and growth.

This is another very interesting thought as well. I've always thought that joining a place that's struggling or is chaotic is a great career and skill builder for proactive and entrepreneurial minded people (I categorize myself in this group). You can be part of the team that turns things around by taking bigger but calculated risks and creating structure from nothing. Really fun work as well. I essentially work at a place like this now (more on the chaotic just starting up end not struggling end) and it's been a great learning experience so far.

But then I also think that there's a difference btw a "sinking ship" and a "damaged ship" that can be repaired. Where would you draw the line?

prikolno


Total Posts: 94
Joined: Jul 2018
 
Posted: 2021-03-27 14:57
I meant the overall culture, recruiting process and workflow. I personally don't like namedropping firms out of respect for friends, former colleagues or business acquaintances who own those firms or manage teams at those firms, so I won't be very specific about the who and why.

Being hung on speed doesn't necessarily imply less focus on problem solving skills. However, some firms do handle more structural relationships in their trades or have a semi-discretionary approach, while others rely heavily on simulation and signal-based alphas, and this usually has some correlation with the amount of rigor applied.

> That's interesting. What would you say are key qualities you and other resume reviewers are looking at?

Probably too many details to exhaust in a short post. There's a fuzzy, conditional logic when you parse a resume, over time you just gain an intuition for it and it takes maybe 8-15 seconds to scan and reject a resume manually. Some examples:

- There are about 110+ financial firms and 20+ tech firms for which the candidate warrants additional attention if they've worked at any of them.
- There are certain non-financial, but work-acquired skills or experiences that are translatable, and that the particular hiring firm has more luck monetizing. For some firms, it could be bioinformatics; for others, it could be game development or audio processing.
- The neighbors of the keywords listed in the job description are more informative than the ones actually listed in the job description, and spotting the neighbors is an intuition you gain from recruiting. An inexperienced candidate usually just lists superficial keywords. If template metaprogramming or Boost are explicitly listed in the job posting, it is obviously more informative that the resume says "Standards Committee".
- Specific groups or associations, e.g. Dave Donoho's group. Erik Demaine's group. Emory REU. CWI.

Undergraduate program affiliation is usually more informative in China or India. Going to a top college or university in the US or UK can often be a negative signal.
Previous Thread :: Next Thread 
Page 1 of 1