To be a quant, which area should I choose for Math PhD study?

[RafaelPrince]

Hello,

Currently I am pursing my math PhD degree and will decide which area to choose in coming months. My career goal is to enter a top-tier hedge fund (e.g. D E Shaw) as a quantitative analyst and work on trading strategies. It would also be great if I can enter an asset management department of a bulge bracket bank. Since my understanding of hunting jobs is very limited, please help me decide which area to choose and what I should learn in the coming years.

Generally I have three choices for my PhD: probability/stochastic processes, PDE, discrete mathematics(related to computer science and might be good for programming skills). I am also considering to pursue a master degree of CS or statistics at the same time. Which master degree do you recommend? And which of the three do you think is the most important knowledge/skills for a quant: programming/CS, probability/stochastic processes, PDE.

Since my sophomore year, I have been focusing on improving my academic background, so I have never had any internship experience. Ideally PhD student get into a summer intern program after the 4th year and find a full time position after graduation. But I wonder whether it will be greatly beneficial for my career if I try to have some internship experience at the early stage of my PhD study. I know it might be difficult since employers usually prefer students who will graduate soon after the intern.

Your advice will be very helpful for me. Thank you very much!

 

[Daniel Duffy]

a simultaneous PhD/MSc

?

 

[RafaelPrince]

a simultaneous PhD/MSc

?

Yes. In my university, a PhD student can request to pursue a master degree in a different discipline at the same time.

 

[Daniel Duffy]

Yes. In my university, a PhD student can request to pursue a master degree in a different discipline at the same time.

That's cool. Do you have to attend lots of lectures? Is there time for research?

I think a CS foundation for math/finance does no harm. But don't forget quality time.

 

[RafaelPrince]

That's cool. Do you have to attend lots of lectures? Is there time for research?

I don't know since I haven't tried yet. You see that I am still considering which area of research to focus on. Any suggestion?

 

[Daniel Duffy]

I don't know since I haven't tried yet. You see that I am still considering which area of research to focus on. Any suggestion?

Hard to say in general. If I were a young lad again and I wanted future proofness I would try to become good in

(applied) maths
CS (data structures and algos)
Statistics A-Z
Become a good programmer (Python, C, C++)
Macroeconomics

 

[pingu]

People are going to see you as a professional student. You will spend an inordinate amount of time in school without getting any real experience.

 

[Daniel Duffy]

In the UK and Ireland you can(could?) get a PhD 3 years after BA/BSc, once you get that brilliant fundamental result. On mainland Europe it tends to be ~ 4-5 year. AFAIK the States is 5-6 years?(?)

 

[Philip_Mitterrand]

There is lot of hand holding in States unlike EU/UK where you are just by yourself.

In the UK and Ireland you can(could?) get a PhD 3 years after BA/BSc, once you get that brilliant fundamental result. On mainland Europe it tends to be ~ 4-5 year. AFAIK the States is 5-6 years?(?)

 

[MartinOvie]

People are going to see you as a professional student. You will spend an inordinate amount of time in school without getting any real experience.

True. But how does a professional student get that experience?

 

[Daniel Duffy]

There is lot of hand holding in States unlike EU/UK where you are just by yourself.

You can say that again. For both MSc and PhD my supervisor threw an article at me and said "Generalise it".

 

[RafaelPrince]

What areas of pure mathematics research are best for a post-PhD transition to industry?
I found this link, which said probability should be the best area to choose in this case. Do you agree?

 

[TehRaio]

probability and statistics have lots of application, so yeah it's kind of obvious.
then again, I won't be the first one to say it: Don't do a PhD just to get a job as a financial engineer, it does not make sense. Work for a few year, and do a MFE or don't do it, you may be able to get your foot in somehow.

 

[Jakelaker]

Hard to say in general. If I were a young lad again and I wanted future proofness I would try to become good in

(applied) maths
CS (data structures and algos)
Statistics A-Z
Become a good programmer (Python, C, C++)
Macroeconomics

Why study macroeconomics?

 

[yetanotherquant]

PhD is - unless you are genuinely interested in science - is a waste of time.
Deep math knowledge will not hurt but the quantitative trading strategies are not about complicated math. Forget millions of papers that (implicitely) affirm the opposite - they are written by virgins, lecturing on sex since these virgins need publications for their academic career. Rather search for contact to those, who make money on the market. It is not easy (both to make money by trading and to get in touch with such people but if you succeed, you will likely see that their models are not too mathematically complicated). Quantitative trading is mostly about data analysis (which includes acquisition, screening, clearing) and programming.

 

[amraj]

PhD is - unless you are genuinely interested in science - is a waste of time.
Deep math knowledge will not hurt but the quantitative trading strategies are not about complicated math. Forget millions of papers that (implicitely) affirm the opposite - they are written by virgins, lecturing on sex since these virgins need publications for their academic career. Rather search for contact to those, who make money on the market. It is not easy (both to make money by trading and to get in touch with such people but if you succeed, you will likely see that their models are not too mathematically complicated). Quantitative trading is mostly about data analysis (which includes acquisition, screening, clearing) and programming.

I’ve been searching the London job market recently, and although its very sad to see, this is not true. Almost all roles advertised at Hedge funds require a PHD nowadays. They seem to want experts in focused areas like Machine learning and cluster analysis. I don’t agree with this approach on hiring, because other skills are becoming increasingly important in a world where everything must be made transparent. You need to be able to meet with clients and regulators to explain complex ideas. Also, you might find that some MSc students are sharper and more creative than someone with a PHD. It appears that the only route into a hedge fund is to bring your own strategies which have been successful (unlikely for a graduate), or to have the PHD. Regardless of this, I wouldn’t advise anybody to do the PHD just to get a role with the funds, because you will regret it.

 

[Philip_Mitterrand]

I have never seen someone walking into HF without quality exp at a bank or from a tough selective program. You need to pass the filtration before even being a contender.

I’ve been searching the London job market recently, and although its very sad to see, this is not true. Almost all roles advertised at Hedge funds require a PHD nowadays. They seem to want experts in focused areas like Machine learning and cluster analysis. I don’t agree with this approach on hiring, because other skills are becoming increasingly important in a world where everything must be made transparent. You need to be able to meet with clients and regulators to explain complex ideas. Also, you might find that some MSc students are sharper and more creative than someone with a PHD. It appears that the only route into a hedge fund is to bring your own strategies which have been successful (unlikely for a graduate), or to have the PHD. Regardless of this, I wouldn’t advise anybody to do the PHD just to get a role with the funds, because you will regret it.

 

[yetanotherquant]

I’ve been searching the London job market recently, and although its very sad to see, this is not true. Almost all roles advertised at Hedge funds require a PHD nowadays. They seem to want experts in focused areas like Machine learning and cluster analysis.

What they write in Job/Ads and what they really require may be different .
Disclaimer: I am not an expert for English Hedge Fond Labour Market, but I know enough guys that work "merely" with German Diplom or Msc in banks, funds and proprietary trading companies.

 

[zyz]

If you are already in a PhD program, then don't listen to those who are saying its useless. It is useful. But you will have to know how to make it useful.
Obviously there are different types of HFs and different types of jobs. Keep monitoring the job market and develop the required skills.

 

[yetanotherquant]

If you are already in a PhD program, then don't listen to those who are saying its useless. It is useful. But you will have to know how to make it useful.

Nobody (in either case me not) says that Ph.D. is useless. I say that unless a person has genuine interest in science a Ph.D. is a waste of time. Indeed, wasting 5 years of time (instead of getting professional experience) and ending-up with a mediocre dissertation will not thrust a career.