Math\Statistics Profile Evaluation

[phd_aspirant94]

DEGREES & GPA: 

• MSc Mathematical & Computational Finance, University of Oxford. GPA: equivalent to 4.0, graduated joint first in the cohort
• MSc Mathematics, Minor in CS, ~top 10 German university. GPA equivalent to 4.0, full marks in every exam
• BSc Mathematics, Minor in Economics, ~top 10 German university. GPA equivalent to 3.8, 3rd in cohort of 50. 

RELEVANT COURSEWORK ACROSS ALL DEGREES:

Linear Algebra, Advanced Linear Algebra, Algebra, Real Analysis, Complex Analysis, Fourier Analysis, Functional Analysis, Probability Theory (including measure theory), Statistics, Optimization, Numerical Analysis, ODEs, PDEs, Scientific Computing, Machine Learning, Reinforcement Learning, Stochastic Calculus, Stochastic Control, several courses in financial mathematics (stochastic volatility, mathematics of fixed income markets, numerical methods for option pricing), Monte Carlo Methods, Time Series Analysis, and some more.

RESEARCH EXPERIENCE:

• MSc thesis Oxford: Worked on machine learning applications to risk management. No publication yet, but am preparing one with my professor. Got awarded a Europe wide price for the best MSc thesis in quant finance selected from 13 European quant finance programs. Will get strong letter from supervisor. 
• MSc thesis Germany: Thesis about Levy processes in financial math. Developed my own financial model, compared with calibration performance of standard methods. No publication from this but will get very strong letter from supervisor. 
• Several other smaller uni projects probably not worth mentioning

RELEVANT WORK EXPERIENCE:

• Worked for two years as a quant researcher in a top tier prop trading firm. I don't know if admission committees would consider this research experience even though it technically is. I am not allowed to talk about specifics due to confidentiality. 
• Two other internships in quantitative risk management.

GRE:

Not taken as not required by most schools. Might take one end of November for programs with deadlines in January. 

I am planning to apply to PhDs in statistics, mathematics and operations research which have groups focussing on financial applications:

• Harvard Statistics
• Princeton ORFE
• Columbia IEOR
• Columbia Statistics
• UChicago Computational and Applied Maths
• UChicago Statistics
• Yale Applied Math (also considering stats and data science)
• NYU Courant
• CMU Math
• Stanford MS&E
• Berkely IEOR
• Cornell ORIE

I am not sure whether my profile is strong enough for some of these so would be thankful for opinions. 

 

[Stat Assistant Professor]

Your profile looks strong but I think your list of schools is quite top-heavy. Those are extremely competitive programs. I'm not sure how your institution is viewed relative to the most renowned Germany schools like LMU Munich, TUM, or Heidelberg. But European undergrad programs in mathematics tend to be very rigorous in general (with PhD-level graduate coursework in the U.S. being undergrad/Masters classes at European schools), so I don't think you will run into much of a problem with adcoms being concerned about your preparation (which is sometimes an issue for American applicants). 

However, since the competition is so stiff, I would recommend trimming your current list a little bit and adding a few "safer" options. Of the schools on your current list, I do think that you have a chance at Cornell OFIE (given what I know about some of the PhD graduates from that department). You might have a chance at the Stat programs on your list, but because admissions is so competitive, I would recommend adding more "safe" choices.  

Edited October 25, 2020 by Stat Assistant Professor

[phd_aspirant94]

Thanks for your response! I know some other graduates from my MSc with slightly worse profiles who got into Stanford MS&E and Cornell ORIE so I think I got a shot at those. How would you gauge my chances at the Princeton ORFE (which is my favourite program) and the math programs like the ones at CMU and Yale? What "safer" choices would you recommend?

[phd_aspirant94]

Also, I didn't graduate from one of the three German universities you listed if that makes a difference in your assessment. 

[Stat Assistant Professor]

I think the MS from the University of Oxford (a top 5 university in the world) will definitely make a huge difference. That said, I'm just not sure how your profile compares to those who are accepted to Princeton, NYU Courant, and Chicago CAM. It seems like beyond prestigious pedigree and strong academic performance, research experience is much more common these days (for the top programs, anyway). You could take a look at some of the PhD students at Princeton ORFE and Courant and see what their CV's look like... look at the year they matriculated and see if they have any papers that were published in or before their first 1-2 years of study and if these papers were co-written with people from outside their program (if that's the case, then that means they likely had a paper accepted *before* they were enrolled in the program).

I think your chances are above average at many schools on your list, including Cornell ORIE, but I'm just concerned that your list is a bit too top-heavy. And you don't want to end up having zero acceptances. Maybe replace two or three of the schools on your current list with some lower ranked programs. 

[phd_aspirant94]

I did exactly what you suggested. It seems that there are quite a few students at Princeton ORFE who had a paper published before they began the program but equally many who "just" had a MSc thesis like I do. I am aware that I am reaching for the top here but if I am not accepted to one of the top programs I'd rather get my PhD in Europe anyways. I am planning to apply at Oxford, Cambridge and ETH as well. If these also don't work I probably won't have any problems finding a place at a top German university. Do you think I should aim to at least write up my thesis as a paper and put it on arXiv before the handing in the applications?

[Stat Assistant Professor]

18 minutes ago, phd_aspirant94 said:

I did exactly what you suggested. It seems that there are quite a few students at Princeton ORFE who had a paper published before they began the program but equally many who "just" had a MSc thesis like I do. I am aware that I am reaching for the top here but if I am not accepted to one of the top programs I'd rather get my PhD in Europe anyways. I am planning to apply at Oxford, Cambridge and ETH as well. If these also don't work I probably won't have any problems finding a place at a top German university. Do you think I should aim to at least write up my thesis as a paper and put it on arXiv before the handing in the applications?

Yes, you could do that (put up a paper on arXiv). I don't know how closely the adcom members will really read it (there's simply not enough time for that). But you can then list on your CV that you have a preprint (though judging, from your original post, it seems like you have another paper already "in preparation"). You could list this as another paper "in preparation." 

In any case, your recommendation letters will carry a lot more weight than having some preprints (these don't really "count" as much as accepted, peer-reviewed manuscripts... but for early career PhD students, postdocs, and Assistant professors who don't have as deep CVs, it is often helpful to list preprints on your CV). CV and preprints won't carry as much weight as letters of recommendation, though. You should talk with your letter writers and make sure that they can attest that your contribution to the research/manuscripts was primarily your own work and that you have the strong potential to become an independent researcher in the future. If your industry experience also entailed research (as in reading academic papers and coming up with new methodologies/algorithms based on these papers), you can also ask your letter writers to mention that through your work experience, you have the maturity to read and understand complex papers/

[phd_aspirant94]

Thanks a lot for your detailed answers! That definitely helps. I will try to put something on arXiv but am not sure whether I can make it on time. My work definitely requires a great amount of research. I am reading papers on a daily basis and impleting the core ideas and combining them with my own insights to make the models work in our trading environment. Will make sure that my recommenders mention this as well then.

One more thing I would like to know is how important the SOP is, especially the part where I outline my research interest. For me it would be fine to work on anything broadly related to financial mathematics, stochastic calculus, statistical time series modelling, machine learning etc. as long as it is mathematically challenging. Do I need to pin down a specific topic and outline my research proposal while citing relevant papers or would it be fine to broadly discuss my interest?

[Stat Assistant Professor]

3 hours ago, phd_aspirant94 said:

Thanks a lot for your detailed answers! That definitely helps. I will try to put something on arXiv but am not sure whether I can make it on time. My work definitely requires a great amount of research. I am reading papers on a daily basis and impleting the core ideas and combining them with my own insights to make the models work in our trading environment. Will make sure that my recommenders mention this as well then.

One more thing I would like to know is how important the SOP is, especially the part where I outline my research interest. For me it would be fine to work on anything broadly related to financial mathematics, stochastic calculus, statistical time series modelling, machine learning etc. as long as it is mathematically challenging. Do I need to pin down a specific topic and outline my research proposal while citing relevant papers or would it be fine to broadly discuss my interest?

It's probably okay to talk about your research interests broadly. Since you would not be accepted into a specific PI's lab in these programs, you don't need to have a thesis topic -- or even a specific research subfield -- in mind. You can talk about what you found interesting about the papers you read on a daily basis to convey that you do research in your current capacity, as well as your research on applications of ML to risk management and your MS thesis research. You could talk about the directions you are interested in exploring. Since you are preparing a paper for submission, you can talk about that too. 

The letters will matter a lot more though and should highlight your research *potential*. I think your background is pretty strong. Based on your description, your undergrad pedigree probably enjoys a decent global reputation (e.g. in the QS rankings), and your strong Masters performance from Oxford will definitely be a huge boost to your application. I think you should be in very good shape with good letters, but the very top programs (e.g. Princeton, Courant) might be a very tough to crack (since there will be other applicants who *also* have the pedigree, excellent grades, and strong letters, but more research experience/academic publications than you). Your application would likely be "in the discussion" though.

Edited October 25, 2020 by Stat Assistant Professor

[DanielWarlock]

On 10/24/2020 at 3:05 PM, phd_aspirant94 said:

DEGREES & GPA: 

• MSc Mathematical & Computational Finance, University of Oxford. GPA: equivalent to 4.0, graduated joint first in the cohort
• MSc Mathematics, Minor in CS, ~top 10 German university. GPA equivalent to 4.0, full marks in every exam
• BSc Mathematics, Minor in Economics, ~top 10 German university. GPA equivalent to 3.8, 3rd in cohort of 50. 

RELEVANT COURSEWORK ACROSS ALL DEGREES:

Linear Algebra, Advanced Linear Algebra, Algebra, Real Analysis, Complex Analysis, Fourier Analysis, Functional Analysis, Probability Theory (including measure theory), Statistics, Optimization, Numerical Analysis, ODEs, PDEs, Scientific Computing, Machine Learning, Reinforcement Learning, Stochastic Calculus, Stochastic Control, several courses in financial mathematics (stochastic volatility, mathematics of fixed income markets, numerical methods for option pricing), Monte Carlo Methods, Time Series Analysis, and some more.

RESEARCH EXPERIENCE:

• MSc thesis Oxford: Worked on machine learning applications to risk management. No publication yet, but am preparing one with my professor. Got awarded a Europe wide price for the best MSc thesis in quant finance selected from 13 European quant finance programs. Will get strong letter from supervisor. 
• MSc thesis Germany: Thesis about Levy processes in financial math. Developed my own financial model, compared with calibration performance of standard methods. No publication from this but will get very strong letter from supervisor. 
• Several other smaller uni projects probably not worth mentioning

RELEVANT WORK EXPERIENCE:

• Worked for two years as a quant researcher in a top tier prop trading firm. I don't know if admission committees would consider this research experience even though it technically is. I am not allowed to talk about specifics due to confidentiality. 
• Two other internships in quantitative risk management.

GRE:

Not taken as not required by most schools. Might take one end of November for programs with deadlines in January. 

I am planning to apply to PhDs in statistics, mathematics and operations research which have groups focussing on financial applications:

• Harvard Statistics
• Princeton ORFE
• Columbia IEOR
• Columbia Statistics
• UChicago Computational and Applied Maths
• UChicago Statistics
• Yale Applied Math (also considering stats and data science)
• NYU Courant
• CMU Math
• Stanford MS&E
• Berkely IEOR
• Cornell ORIE

I am not sure whether my profile is strong enough for some of these so would be thankful for opinions. 

 

You have an excellent shot. I personally am studying at the Harvard program and one of my best friends in college studied at Princeton ORFE. I don't think my friend's or my profile is much stronger than yours. The only thing is both of us get genuinely strong letters (with preprints) and nearly full GPA like you. If your letter is as strong as you say (looks like it will be), then you can crack these top programs. 

[phd_aspirant94]

Thanks a lot for weighing in your opinion. That sounds encouraging! I'd like to also get your opinion on my LOR selection:

- Oxford prof who supervised the thesis I won the Europe wide thesis prize for. He said that he'll write a strong letter and I am currently writing up my thesis with him. 

- German prof who supervised my first MSc thesis. He is probably not that well known internationally but I am sure that he will write a very strong letter. His letter already got me into the Oxford MSc program. Plus he got his PhD from Harvard so he might know some people as well.

- Oxford prof who was in the commitee for the thesis prize which means he read my thesis. He is very well known internationally but did not teach me in class. He said that he can write about the fact that I graduated joint first and my thesis. However, he mentioned that his letter may be less informative than a letter from someone who taught me in class. I am not sure whether I should take his letter. His name is really big, and his letter will probably be pretty positive but I still have slight doubts. 

Some programs allow up to 6 LORs. Hence, I think of adding another one:

- Oxford prof who taught two classes where I got excellent marks. He moved to a German university in the meantime and offered me to do a PhD with him so I presume his letter will also be positive. 

For the programs where I can only hand in three letters I am not sure whether to select the third of fourth recommender in my list above. What would you recommend? Also, would you try to get all letters from Oxford profs or do you think it's better to als take the German? I believe it's one of the most detailed ones (together with the one from my Oxford supervisor).