PhD Dilemma: Which program to choose?

Joined
2/22/21
Messages
11
Points
13
Hey all,

I am in quite a tough situation having to pick amongst quite a few very good Ph.D. programs:

- Princeton Operations Research and Financial Engineering (with Gordon Wu Fellowship, the highest fellowship in engineering)
- Stanford Management Science and Engineering (with SG Fellowship in Science and Engineering)
- Columbia Industrial Engineering and Operations Research (with Presidential Fellowship, the highest fellowship in engineering)
- Cornell Operations Research and Information Engineering
- CMU Mathematics with focus on Stochastics and Financial Math
- Yale Statistics and Data Science

I am also waiting to hear back from the DPhil in Mathematics at Oxford.

At the moment my two favorites are Princeton and Stanford, which are especially attractive due to the funding. I was wondering whether there is any consensus on what the best program amongst the above is. At the moment I am slightly leaning towards Princeton because there are more people working in mathematical finance and the program seems to be a bit more math heavy. On the other hand, I feel that the Stanford brand is a bit stronger.

A bit about my background and aspirations: I have worked in the industry as a quant for two years now at a tier 1 prop trading shop. Before that, I obtained my MSc in Mathematical and Computational Finance from Oxford. My goal is to become an absolute top quant at hedge funds like D.E. Shaw or Rentech.
 
Thanks for your input! I agree that the entrepreneur culture at Stanford might be a big plus. I am also playing with the thought of interning at a place like Google Brain or MS Research during the PhD, for which Stanford is the better place as well. On the other hand, I am slightly worried about the words "Management Science" in the name of the program. The professors I would work with are all heavily focusing on quantitative work and are all affiliated with the ICME but I am worried that for someone on the outside it might look more like a management degree rather than a heavily quantitative degree.
 
idk where your worry comes from... ppl know management science and operations research are more or less equivalent and surely quantitative

i would personally recommend to study stuff more behavior based and start a shop investing in the "psyche" of the market. i assume you've been paying attention to the market based on your work background. you probably already realized that investing in stuff like tech (covid bet), bitcoin (inflation and gold replacement bet), reddit bro stocks (isolation boredom and peasant uprising), and anything power/gas related (texas power shortage) are making much much more money than any quantitative strategies out there. the funny part is that none of those bet require fast execution and you can be late 1 week later and still pocket a ton
 
Looking back I wish I had applied to ICME. Back when I prepared the apps and was looking for the right program I chose MS&E because most of the professors working in quant finance and stochastics (Giesecke, Blanchet, Glynn, Pelger,...) are MS&E professors. However, all of them are also affiliated with ICME and so I might have been better off choosing ICME. Do you know if it is possible to switch the program?
 
Looking back I wish I had applied to ICME. Back when I prepared the apps and was looking for the right program I chose MS&E because most of the professors working in quant finance and stochastics (Giesecke, Blanchet, Glynn, Pelger,...) are MS&E professors. However, all of them are also affiliated with ICME and so I might have been better off choosing ICME. Do you know if it is possible to switch the program?
Maybe contact the first person who responded on here? How does Stanford ICME compare to Management Science & Engineering? - Quora
 
In the meantime I have also received an offer from the Oxford mathematics department. As Oxford arguably has the best quant finance group worldwide, I am very tempted by their offer. On the other hand I have already studied at Oxford and plan on transitioning to the US long term. The benefits of Oxford would be that 1) it’s a Math PhD as opposed to an OR/statistics PhD, 2) the PhD is only 3-4 years, 3) the courses one has to take at Oxford are much more specialised and advanced than the ones at the US unis and 4) Oxford arguably has the best and biggest quant finance group. The benefits of a US program would be that 1) I can put a different elite institution on my CV, 2) I can probably find a job in the US afterwards more easily and 3) have better funding. Does anyone of you have any opinion on this? Also I‘d be interested in which uni is considered more prestigious globally.
 
i would personally recommend to study stuff more behavior based and start a shop investing in the "psyche" of the market. i assume you've been paying attention to the market based on your work background. you probably already realized that investing in stuff like tech (covid bet), bitcoin (inflation and gold replacement bet), reddit bro stocks (isolation boredom and peasant uprising), and anything power/gas related (texas power shortage) are making much much more money than any quantitative strategies out there. the funny part is that none of those bet require fast execution and you can be late 1 week later and still pocket a ton
can’t say i disagree, quant isn’t worth doing if your only goal is to make a buttload of money. especially when tech is right next door too with higher EV
 
@TehRaio I still didn’t decide. I’ve talked to many professors and students from Oxford, Stanford and Princeton and all these programs seem amazing. Overall, I like the Oxford program the most. However, I think it might be better to attend a different uni and when it comes to working in the US, it might be harder to transition from the UK.
 
"A PhD is more of a character test than intelligence test"
Dr. Sinead Kane

(visually impaired ultramarathon athlete, double PhD doctorate, and double Guinness World Record holder, Dr Sinead Kane.).
 
Am I the only one that thinks this post is crazy? Why are you trying to do a PhD. If you're already at a "top" fund, spending 4-5 years in a PhD program isn't going to do anything to help you become "an absolute top quant".

Don't go to any of these programs if that is your goal, just stay at the fund you're at and learn whatever you want to learn in your free time.
 
There’s several reasons. First, it was always my dream to get a PhD. Since the day I started my job as a quant I knew that I wanted to return to academia to get my PhD. Secondly, I firmly believe that you cannot build the skill set you develop during a PhD if you’re working full time and only learn new stuff in your spare time. Many of my daily duties don’t involve any complicated math, and so I think building the skill set I want to have on the job is much harder than to build it by pursuing a PhD. Finally, I’d also like to broaden my quantitative skillset beyond the skills that are strictly required to be a good quant.
 
Am I the only one that thinks this post is crazy? Why are you trying to do a PhD. If you're already at a "top" fund, spending 4-5 years in a PhD program isn't going to do anything to help you become "an absolute top quant".

Don't go to any of these programs if that is your goal, just stay at the fund you're at and learn whatever you want to learn in your free time.
It doesn't work like that.
 
There’s several reasons. First, it was always my dream to get a PhD. Since the day I started my job as a quant I knew that I wanted to return to academia to get my PhD. Secondly, I firmly believe that you cannot build the skill set you develop during a PhD if you’re working full time and only learn new stuff in your spare time. Many of my daily duties don’t involve any complicated math, and so I think building the skill set I want to have on the job is much harder than to build it by pursuing a PhD. Finally, I’d also like to broaden my quantitative skillset beyond the skills that are strictly required to be a good quant.

I have a PhD in one of the fields you've listed above (albeit not from one of those schools, but from a pretty darn good one), however, I don't have experience working as a quant, so take what I'm about to say (especially on the quant side) with a grain of salt.

First, if indeed your goal is to just become "an absolute top quant" getting spending 4-5 years getting a PhD is likely not worth your time and is not going to help you achieve this goal with any greater efficiency or efficacy than just staying in your current position.

Now, that said, it seems like, despite your username and stated goals, you do have other motivations for getting a PhD. So i'll give my opinion of reasons you should (or should not) get a PhD. I should say "you" here is very specifically the OP, I would not necessarily give this same advice to a general person trying to decide if they should pursue a PhD or not.

You should get a PhD if any or multiple of these apply -
1. You don't mind forgoing 4-5 years of (probably pretty high) income and experience for the sake of learning and publishing research
2. You will always feel like something in your life missing or that there will be a nagging feeling of "I could've done more" if you don't get it
3. You have personally experienced your career being stifled because you lacked a PhD
4. You really care about and are interested in the subjects you will potentially study, and as such are willing to trade your (probably pretty sweet) current situation to rough it as a graduate student for 4-5 years for the sake of personal knowledge and growth.
5. You will still be happy in your PhD if your adviser leaves the school and you must change topics. You will still be happy if your adviser suddenly dies (unlikely but it happens). You will still be happy when the paper you've been working on for 2 years gets rejected because some reviewer didn't think your proof strategy was novel enough.
6. You think you may want to be in Academia instead (although this is a whole other discussion itself)

You should not get a PhD if any or multiple of these apply -
1. You think a PhD will get you more money but you have no direct evidence that this is the case
2. You think that the PhD will confer you special skills that you don't already have. The purpose of a PhD is not to contribute to the overall knowledge of humanity, despite whatever romanticisms you may hear. The purpose of a PhD is to teach you how to do research. To teach you how to take on a previously unsolved problem and solve it. Ask yourself - do you truly feel like you do not already possess this capability? Frankly, if you're already working at a top fund your ability to do research is likely "good enough" and may even be better than a lot of PhD graduates.
3. You think that getting a PhD will allow you to be a "different type" of quant that sits around proving theorems all day and doing more hardcore math. Maybe this is true (as I mentioned, i'm not in the industry...yet) but I highly doubt it. Even if such positions did exist - shops are filled to the brim with top Math talent from HYPSM schools that don't do this - so what makes you think you are so special that you would?

EDIT: I should add that if I were you, OP, I actually probably would take the PhD - but that's because i'd answer affirmatively to all of the questions I listed. However, it's your stated initial motivations that make me recommend you don't do it.
 
Thanks for your detailed response! You’re right in that I have other motivations to pursue a PhD. I would definitely answer yes to most of the questions you asked (except for no 3). My situation is more that I want to pursue a PhD because of the reasons you stated. But if I then have the choice between several programs I might as well choose the one that in addition to my academic goals also has the most beneficial side effect on my career prospects.

Since you already have a PhD, I’d be interested in your advice on the choice of the program.
 
Back
Top Bottom