Physical Science PhD to Finance?

[Brenton]

Hi everyone,

I'm new here and had some questions about the finance world. I'm currently in a Chemistry PhD program even though I don't really do chemistry -- most of my work is in applied mathematics/stochastic calculus/probability theory towards biological models. I've begun to start looking for opportunities after graduation to continue using my math knowledge as a career. I originally wanted to stay in academia but I wanted to be a math professor, and since I'm getting a chemistry PhD, I'm not sure how likely that will end up being, so I'm looking at possible alternatives. Everywhere I look, it seems that finance is the main field that uses stochastic calculus/advanced math topics, so I've begun looking into this as a career. I'll be honest that I don't know much about finance but I do know a lot of math so that may be a problem (I know some people who did econ/math double majors in college and got a job on wall street right after graduating). I've thought about taking the actuary exams as proof that I can do math and it would force me to learn more finance terminology and facts. A little about me and my background:

Bachelor's in Chemistry and Math
Undergrad GPA: 3.67

Master's in Chemistry
Currently working on PhD in Chemistry (possibly with Applied Math specialization)
GPA: 3.88

I do some Python coding for my research though most of the math I do is more analytical than numerical, so computational work is really secondary to the analytical work I do.

I'm not sure if I would even stand a chance of entering this field, but I feel like my math skills would certainly help, and not many people in the finance field have physical science backgrounds which would help me stand out I suppose. Or, perhaps there are other fields I should look into where I can use my knowledge of stochastic processes in other ways? Anyways, I've only started looking into this field a few weeks ago so I'm trying to learn as much as I can about the quant/finance world!

 

[C S]

"I feel like my math skills would certainly help, and not many people in the finance field have physical science backgrounds which would help me stand out I suppose"

Not many people doing straight-out accounting-type finance have physical science backgrounds. They have (duh?) finance/accounting backgrounds. But there are many types of jobs in finance, and certainly there are plenty of physical science types, and I'm not even talking about the quant space.

"I do some Python coding for my research though most of the math I do is more analytical than numerical, so computational work is really secondary to the analytical work I do."

I'd learn more Python. It will be of immense benefit to you when you start searching for jobs, on and off Wall St.

 

[Liam]

Hi everyone,

I've thought about taking the actuary exams as proof that I can do math and it would force me to learn more finance terminology and facts.

Not sure about US but I know a leading probabilist that literally had to rewrite the entire course when he took over the stochastic processes course from an actuary in his department. And actuaries I know say this isn't isolated - most exams in UK and Ireland are copy and pastes from notes and some actuarial notes come out with vomit inducing crap like "as we're differentiating wrt to a random variable we can't use the usual chain rule. But if we pretend we can we get the right answer". If it's the same in US then interviewers will know about it and will majorly give more than just 2 flying fucks. The exam thing in UK actuary is a farce.

It's a nice idea, but not an actually good one, especially as it'll take ages and, duh?, make it look like you want to be an actuary, a vastly different field. Why not read a few derivatives books like Hull? Nothing wrong with looking at actuary - a much less stressful field, but don't confuse it with quant finance.

As the previous poster suggests, watch out with the made up as you go along reasoning. Not having a typical quant background is also something an English major has, so why not teach one that understands basic probability a little more than most a bit of stochastics? But you have a PhD background - although MFE is ideal now, I think many employers would rather employ PhDs as it is a very practical degree and it proves you can take charge of a problem. Although you will find some need for brushing up technical things (the prep is actually dog rough), selling that last point is probably more important - simply translate what you did in phd into usefulness in finance.

The only concern I would have isn't technical - it is that this mission to get a job using stochastic processes will distract from finishing the phd. Sort that out, THEN polish up your skills. I suspect you're like I was when I finished uni - believing what others around me said ie that you're great and will never fail, blah blah blah, but I was lucky enough to have one of my phd housemates advise me to tone down the stupid and useless "job hunt" I was engaging as he warned that, although my grades were good, he'd seen people in my position fail from the distraction and stress of jobhunting.

Not only did my grades improve after taking that decision to focus but I found my feet after graduating and, as he predicted, some others I was in uni with actually wound up with no msc or phd thanks to these kind of distractions, a deadly hole in their cv, even though they'd probably never previously failed an exam in their life and were strong at maths. Worth thinking about.

 

[pingu]

What's your PhD topic? You mentioned every ancillary thing except your topic.

 

[cab]

https://www.quantnet.com/threads/mathematical-biology-to-quant-finance.19915/#post-149749

 

[Daniel Duffy]

"Or, perhaps there are other fields I should look into where I can use my knowledge of stochastic processes in other ways?"

I think this is the wrong focus. An MSc would have done just as good. A PhD means you can think independently (research). The subject matter is irrelevant.

Actually, stochastics is quite a small part of computational finance. In university, you tend to choose research topics that 'fit' your supervisor's research interests. In industry they tell you to solve a problem ASAP. Are you flexible?

Focus on finishing the PhD. Napoleon and Adolf H. prove you cannot fight on two fronts, simultaneously.

"I'll be honest that I don't know much about finance"
And you want to get into finance? You realize >= 80% of your time will be spent programing a.s.?
The bright lights?

 

[C S]

Maybe this is a bit random, but I'm surprised how many people that come into an interview as master of stochastic calculus get demolished by problems involving a one step binomial tree.

 

[Liam]

Maybe this is a bit random, but I'm surprised how many people that come into an interview as master of stochastic calculus get demolished by problems involving a one step binomial tree.

Not really - a similar example I was given as a postgrad was removing volatility in the BS sde. And many MFEs won't realise it's now a simple pde (the stock is a riskless one with exponential growth).

makes the difference between what grads think quant is - maths and programming and what it actually is - using whatever helps traders and market risk managers do their job properly. So if it just requires an elementary spreadsheet so be it. The problem with grads sometimes isn't necessarily not reading up the Hull books etc but basics like not knowing when to put their hand up and ask for help in something, or even just to have someone look at what solution they have so far, to avoid hours of going in the wrong direction. T

Nope not a random post at all. OP's gotta sell themselves as practical in that sense.

 

[Liam]

Not really random - these kind of examples are the kind of things that got mentioned to me when I was doing interview preparation. Another one is asking then to solve the BS equation with no volatility. Part of the problem is a quality issue but the issue I have with it isn't technical - it's about willingness to solve the financial problem at hand and realising maths is just a tool.

Personally I think a lot of the culture shocks of entering a corporate world most msc in stochastics students will easily overcome. But 2 are where I see divergence - the example you mentioned is important as often we just used whatever was necessary, it wasn't always necessarily hard maths or complex programming. The other thing not really mentioned is learning to be properly productive ie when to get code or methods sanity checked and admitting to not knowing something.

 

[Daniel Duffy]

Maybe this is a bit random, but I'm surprised how many people that come into an interview as master of stochastic calculus get demolished by problems involving a one step binomial tree.

They will soon learn after getting out of the university, by degrees. Sink or swim.

 

[Brenton]

What's your PhD topic? You mentioned every ancillary thing except your topic.

Whoops! So my PhD topic is on modeling cognition as a dynamical system. Essentially what I'm doing is incorporating features of cognition (e.g. robust dynamics, stimuli sensitivity, good behavior regardless of initial conditions, etc.) and building a topological structure which uses those features (namely a heteroclinic cycle). Then, because neuronal systems are noisy, subjecting dynamics within that topological structure to noise and calculating properties about the noisy dynamics analytically.

Liam:
Yes you are right, it is important to focus on my PhD work before fretting about the financial world. I wasn't sure if there were things I could do now to help prepare myself so I don't spend a few years after graduate school trying to prepare myself then when I might be able to start now.

Daniel Duffy:
Yes I am flexible to other math areas though most of my work is stochastic dynamics.

"And you want to get into finance? You realize >= 80% of your time will be spent programing a.s."

Actually no, I was not aware of this, and one of the reasons for my posting here was to find out what is involved in quant finance. I imagined there to be some programming though I'm unaware of how much people in this field actually do, and how much math people really use on a day to day basis.

I also don't know how much math people in actuarial work do either, but one of the reasons I had considered taking the actuary exams is because I thought there might be fields in the actuarial world where I could apply a lot of the math I've learned and studied. Plus, I thought it might be easier to get a job as an actuary than it would to get a quant job, and I could still do a lot of the math I love.

 

[Daniel Duffy]

I think daily quant work is much different from daily academic activity.

Plus, I thought it might be easier to get a job as an actuary than it would to get a quant job, and I could still do a lot of the math I love.

As Liam said, math is a tool in this context. Why not stay in academia. I am not sure how much hard core math is used in actuarial work??

 

[TheComplexUnit]

Whoops! So my PhD topic is on modeling cognition as a dynamical system. Essentially what I'm doing is incorporating features of cognition (e.g. robust dynamics, stimuli sensitivity, good behavior regardless of initial conditions, etc.) and building a topological structure which uses those features (namely a heteroclinic cycle). Then, because neuronal systems are noisy, subjecting dynamics within that topological structure to noise and calculating properties about the noisy dynamics analytically.

This sounds a lot more interesting than working in finance. Don't be a sell out.

EDIT: the allure of quantitative finance dwindles quickly .