• 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!

Current math PhD

Joined
8/28/07
Messages
7
Points
11
I'm currently a math PhD student, I am doing a thesis on partial differential equations (not stochastic). I still have time to take courses in my department and I was wondering what I should do? I don't think I have a good chance at becoming a professor. I've always liked investing, I invest on my own and I have read many books by famous investors (Lynch, Buffett, etc), so I'm not all of a sudden saying I can't succeed in academia, let me give finance a try. I do enjoy it and would like to make a career out of it.

I did have some questions:

(1) I've heard people say that they want PhD's out of "TOP" universities. What exactly does this means? I am currently going to a top 15 math PhD program, top 10 for my subject, but it's not classically considered one of the "TOP" schools like Harvard or MIT. It's not an Ivy League either.

(2) I currently have very poor programming skills at the moment. If I learn a lot of C++ and maybe learn C++ from a book geared towards quants, will this be sufficient?

(3) How exactly do you transition from Math PhD to quant? Do I need to do an internship during my PhD? Do I contact a headhunter before I am about to finish my thesis?

(4) Do I HAVE to take courses like stochastic analysis or probability at the grad upper division level? I say have to because I still have time and I can take those courses in my uni, but are there hedge funds out there that are willing to teach it to you?

Thanks!
 
I would really be surprised if hedge funds will teach you stochastic or for that matter any subject. Generally hedge funds look for readily deployable talent. There are very few hedge funds that really invest in training, they will most likely expect you to know such topics. However, if you are in exceptional in PDEs say--and they need that skill they might recruit you and train you on what you already do not know.
 
On number 2: I recommend learning C++ ( or Java, or C# ) from a generic book that teaches you the language. Most quant finance / C++ books I've seen are more useful if you already know how to program. So, get a book that teaches you how to program correctly, and then spend time building models. Any book will teach you fundamentals, but learning how to create a model or solve a problem computationally requires time spent coding and debugging. At least this has been my observation.
 
(3) How exactly do you transition from Math PhD to quant? Do I need to do an internship during my PhD? Do I contact a headhunter before I am about to finish my thesis?

(4) Do I HAVE to take courses like stochastic analysis or probability at the grad upper division level? I say have to because I still have time and I can take those courses in my uni, but are there hedge funds out there that are willing to teach it to you?

Eight or ten years ago it was easier, as a Ph.D. in math plus programming knowledge and experience was enough to get one's foot in the door (there were only five or seven MFE programs at the time and employers were willing to train a little bit). But this situation has obviously changed. There are many experienced people on the street at the moment, able to hit the ground running. You have to compete with them. There are Ph.Ds who go on to do MFEs just for the knowledge and qualification.

Grad upper division level stochastic is probably not what's needed. I'm thinking of books like Revuz and Yor -- that's not what's needed. There are now several excellent stochastic books geared towards finance (Shreve is one example). Any of them should be enough.

No harm in talking to people in the business to get a feel for what's happening, what the possibilities are, what's needed.
 
look on this forum for dominic connor's guide to being a quant, he is a head hunter who has written up what appears to be a rather good answer to most of your questions.

you send him a resume, he sends you his pdf.
 
One of the professors at my university did their thesis under Shreve. I might take some probability, stochastic courses and try to do an independent reading with this professor on Shreve's text.

At this point I'm wondering if I should try to focus more on mathematical finance. My uni has a small math finance group, but like I said, the professors are from well known advisers (I think 2 are from Shreve). It's certainly not Stanford, NYU or CMU, but I think it's a respectable math finance program. It's not too late for me to switch my focus from PDEs to mathematical finance. I would stick with PDEs if I had a good chance at becoming a professor, but that seems very doubtful at this point.

While everyone says becoming a quant is hard, it's also really hard to be a professor. A lot of the professors at the top 30-40 math schools are from the cream of the crop. Getting a good postdoc, getting a tenure track position, etc, it's very hard. I would estimate that easily less than 20% of math PhDs eventually become tenured professors. While being a quant might be harder, maybe has a 10% rate, I don't know, but it seems to be a better job.
 
At this point I'm wondering if I should try to focus more on mathematical finance. My uni has a small math finance group, but like I said, the professors are from well known advisers (I think 2 are from Shreve). It's certainly not Stanford, NYU or CMU, but I think it's a respectable math finance program. It's not too late for me to switch my focus from PDEs to mathematical finance. I would stick with PDEs if I had a good chance at becoming a professor, but that seems very doubtful at this point.

Which university are you talking about?

While everyone says becoming a quant is hard, it's also really hard to be a professor. A lot of the professors at the top 30-40 math schools are from the cream of the crop. Getting a good postdoc, getting a tenure track position, etc, it's very hard. I would estimate that easily less than 20% of math PhDs eventually become tenured professors. While being a quant might be harder, maybe has a 10% rate, I don't know, but it seems to be a better job.

Maybe less, much less than 20% of Ph.D.s go on to become tenured professors. That's why math Ph.D.s (as well as physics Ph.D.s) have been drifting towards quant work). If you're not from MIT/Harvard/Princeton/Berkeley/Chicago and maybe one or two other schools, it can be very hard to find a tenure-track position at a research university. It's hard enough with a Ph.D. even from such ranking schools.

Quant work is insecure -- difficult to even call it a profession. No-one know where it will be five or ten years from now. Also the work is stressful and involves long hours. In contrast, once one has tenure, life is not insecure and not stressful.

It's probably not so much stochastic that will make you a quant as lots and lots of heavy-duty scientific programming. If you're in PDEs, use the opportunity to get in lots of such programming and to learn as much numerical analysis and design of algorithms as you can. Also heed the sage words of Paul Wilmott: "PDEs are forever."
 
Back
Top