• TIME TO 2024 UK RANKINGS

  • C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Physics PhD student considering career, need advice

Joined
5/31/14
Messages
1
Points
11
I will be graduating from University at Buffalo SUNY in about 6 months and I'm not certain that I want to stay in physics. At some time in the past it was suggested to me that I consider looking into quantitative finance, so I thought I would ask about it here.

My qualifications are:
Undergraduate degree in physics with a math minor
-Specific classes in computational modeling and game theory
Will have a PhD in experimental physics
Experience modeling nonlinear systems in C++ with some experience of MPI and openMP
Departmental server administrator for 2 years
Proficient in:
-C++
-Java
Basic ability in:
-Python

My questions:
1. Would I be considered qualified/competitive by companies that are recruiting
2. If you have a degree in physics and made the transition, do you enjoy your job?
 
Mathius, rather than joining the "you'll have no problems" brigade I'd ask you to rephrase your question.

All you've done is name fields of study - it's not about that. You need to outline what you can do that's useful within finance - e.g. a friend of mine studied properties of matter during his PHD. As market microstructure problems are solved using similar processes he then easily got a quant job using the same processes he used in his PhD.

That's just an example - I suggest finding where there's a logical connection between your PhD and being a quant - I think this will easily be found e.g. perhaps look at problems with tail risk, where most models fail, but where complex nonlinear models could do better. There is a "simple is better" thing going through finance right now with models, but ultimately there will be companies that will want your skills.

I would pursue it, but remember that most of these people suggesting quant finance probably won't know about or understand the job or risks. Contrary to what sites like this might have you believe the job can be very tough, especially if you get into the wrong team or part of the bank. It's usually brushed under the carpet with some patronising crap about banking "being cutthroat", but it's not really about that - being in the wrong job is very hard to sort out within banking simply because your skills aren't being tapped up and it only takes one bad business decision for you to wind up in the wrong job.

Also its not the fait accompli people think it is - even when I started having a PhD was no guarantee of becoming a quant.
 
The wrong job thing Liam mentioned. I'd really try to understand what he's saying. It happens quite a bit to PhDs, I think.

The reality is that PhDs are kind of an awkward fit now. Banks really aren't trading much like they used to. And a lot of those fancy phd quant jobs you heard about were at big firms that could spend on research and training of academic types. My experience is that a lot of professors are severely out of touch with the job market. They're going on what they heard about some phd grad 10 years ago got. It's 180 from that nowadays.

The typical quant at a bank, even the ones in "research" aren't doing research. They aren't even doing implementation most of the time. The core implementations are too important to be given to somebody with only a few years experience. So most of the time even the hotshot entry level quant is wrestling with some build or deployment or bug fix issue. And you don't need to hire a PhD to do that.
 
So most of the time even the hotshot entry level quant is wrestling with some build or deployment or bug fix issue. And you don't need to hire a PhD to do that.

In Japanese industry, testing is done by the most senior engineers.
 
The reality is that PhDs are kind of an awkward fit now.
Which means some PhDs are actually working in call centres, while graduates with less advanced degrees might find it tough to get a "job" but they'll get something decent eventually. One way out of this is to look at stuff that's hiring - maybe data analysis for a firm implementing Hadoop. But again you run into the awkward fit issue.
My experience is that a lot of professors are severely out of touch with the job market. They're going on what they heard about some phd grad 10 years ago got. It's 180 from that nowadays.
Which is a shame, because they might understand the issues. Thing is explain this to family and friends and they'll shove well meaning but useless advice to "suck it up" down your throat. Thing family, friends and professors even will never get is that employers are the ones that choose - no matter how much you much you want it unless the job fits to perfection employers will not hire you. Family won't understand it and careers advisors will simply state some crap about it being in your mind, while taking your money. Psychologically you're probably fine but to avoid being exposed as frauds most careers advisors come out with this rather than fess up to not understanding your market and the pressures you face. Don't fall for their "Only you can help yourself" crap either - it's not therapy and half the career advisors out there wouldn't know therapy if it hit them in the face.
The typical quant at a bank, even the ones in "research" aren't doing research. They aren't even doing implementation most of the time. The core implementations are too important to be given to somebody with only a few years experience. So most of the time even the hotshot entry level quant is wrestling with some build or deployment or bug fix issue. And you don't need to hire a PhD to do that.
The underlined is the most important bit - likely you'll be unemployed for a while and when explaining this to Dad, or whoever pays for you, ensure you get to the underline bit asap. The rest is fine, but for someone who's bleeding cash for you they'll think you're being picky - it's not like 1960s where people could take gap jobs, it's 180 where employers are suffocatingly picky. Recently I read an article about how for a toilet cleaning job the company hiring refused to interview unless applicants were planning on becoming professional cleaners. It's same with waiting tables to pay bills. I'd seriously doubt lieing in the process or any other ill thought out ideas people will come up with will help. This also links in with how to get a job - find out where you're needed not where there's "loads of jobs" or these industries that are "doing brilliantly" - it's not a case of big degree=whatever you want. I'm actually seeing experienced professionals in some fields begging for clerk work, saying they'll do it for free, orwhatever, and being turned down as the employer simply doesn't want that. E.g. IT is doing brilliantly but thing is outside of data science where do you fit? And don't fall for crap about it being about it just being tough to get on the ladder - even if an industry like that hired you, it's more probable you'll last 2 weeks, or worse still 2-3 years and then unemployable. Even if you really wanted a job that's not real PhD fit, eventually it just won't work out. I'm sure there's plenty of "potato peeling" in whatever the perfect fit is, but it's not about that, it's about playing to your strengths.

Also, again, family and friends will never get the message - I've been shouting and hollering this at my Dad for 10 years now and he still doesn't get that as I've worked in the industry it makes me the expert not him, no matter how arrogant he is or how many shitty newspaper articles he reads. Get used to this - basically until you get a job it's "What would you know, you're unemployed" and people stop short of telling you you're degree is useless. Then when you're employed if you run into issues with work and have very little room to manoeuvre people within the industry will understand. But for family etc it's dismissed, belittled and suddenly people patronisingly announce to you you're degree is useful (as if you not knowing is the reason for issues), now they can use it against you, and conveniently forget that being employed makes you knowledgeable. They completely misunderstand. Or maybe half closing their eyes as if they're saying something profound and speaking in a patronisingly informative tone makes them an expert on whatever they talk about, while all those within my industry don't understand something they've probably been doing for 10-20 years. By the way I'm being sarcastic with that last comment, give it 2 or 3 years and you'll feel the same even if you land a good job.
 
Last edited:
The reality is that PhDs are kind of an awkward fit now.

I suppose less research and more maintenance these days?
 
The reality is that PhDs are kind of an awkward fit now.

I suppose less research and more maintenance these days?
Plenty of research, but generally much more targeted. You are given a regulatory aim and perhaps even a model (e.g. single-factor Vasicek) and you have to implement it times 1 billion. Then you have to parse teh results and examine teh tail very, very closely.
 
Plenty of research, but generally much more targeted. You are given a regulatory aim and perhaps even a model (e.g. single-factor Vasicek) and you have to implement it times 1 billion. Then you have to parse teh results and examine teh tail very, very closely.

In fact, you become the 'owner' of the model A-Z? So make it bullet-proof and stress-tested?

On a general level: I find that taking a single model (e.g. BDT) and working it out in detail gives great insights how to extend it as well discovering what works and what does not work. At least that's the way I tend to approach problems.
 
Last edited:
Back
Top