Statistics/Math PhD/Masters Questions

[Apollo_17]

Hi guys, thank you so much for reading this. I have some questions that I'd really appreciate any insight on. For some background, I was a pre-med until I switched majors as a junior to applied math. I finished all my pre-med courses (2 semesters bio, 2 semesters chem, 2 semesters orgo, 2 semesters physics) and got all A's for reference.

Undergrad Institution: Harvard (Current Senior)
Major:  Applied Mathematics (Focus in Statistics)

Relevant Coursework: Math: Calc 1, 2, 3 (All A's), Linear Algebra (A), ODE/Intro to PDE (A), Complex Analysis (Not proof based, A), Optimization (on track to A), Real Analysis (next semester), Abstract Linear Algebra (next semester), Abstract Algebra or PDEs (next semester).

                                       Stats: Intro to probability (A-), Statistical Inference (A-), Linear Models (On track to A), Machine Learning (B+), Intro Stats (A), Intro Python (A)
GPA: 3.93
GRE: 168Q 164V 4.5 AW

Research Experience: One summer of research in a computational neuroscience lab (no publications). Currently talking to a professor about starting research with them after winter break.

Jobs: I currently work as a TA for a course in Single Variable Calculus, EMS previously

 

My goal is to eventually get a phd in either of Stats/Math/Applied Math at a top school, but I haven't entirely decided on the field yet. I'm very interested in theoretical work, but I haven't really done any proof based courses or research in the area, so perhaps I'm not qualified to make that statement. Currently, my plan has been to get a masters with a thesis in stats at a top school to show I am capable of graduate level work, and to get more research experience, and then apply to PhD programs after. However, after talking to some professors, I am wondering if perhaps that is the best move. I have been advised to consider doing a post-bacc instead of a masters to get that additional research experience needed for admissions, and not risk poor grades. In addition, I've been told that doing poorly in a masters can do more harm than good, so now I am slightly hesitant.

 

Overall, I am wondering if anyone has any insight on whether I should go ahead and do the masters (I am applying this cycle), or wait a year (or two) and try for PhD programs after getting more research experience through a post-bacc or something similar. Additionally, if anyone has any insight on how likely I am to get admitted to any masters or phd programs in my desired fields as of right now, that would be incredibly helpful too. I understand it may be difficult to help me since I have no proof-based courses and am only now taking them next semester, but any help is very appreciated. Thank you for taking the time to read this everyone!!!

[bayessays]

This is mostly a statistics forum, so I'm going to give this advice from that perspective.  You're already in good shape for statistics. (You probably are in decent shape for applied math, but you definitely have some gaps for a math PhD applicant.)

You say you don't have proof-based math courses, but you *will* have Real Analysis after this semester - I'd prioritize getting an A in that, and your math background will be totally fine.  (RA + Linear + Abstract Algebra seems like a heavy load in one semester, so if you have any flexibility, I'd just make sure you can devote the time to do well in RA).

I realize it's already extremely late in this cycle, but you could apply to stats PhD programs right now.  You'd also be a great candidate for top 10 biostatistics PhD programs.  Especially after completing RA and if you're able to start doing research later this year with a Harvard stats prof, I see no reason why you'd need to get a master's before your PhD.  I would just take a year off of school and re-apply.

[Apollo_17]

3 hours ago, bayessays said:

This is mostly a statistics forum, so I'm going to give this advice from that perspective.  You're already in good shape for statistics. (You probably are in decent shape for applied math, but you definitely have some gaps for a math PhD applicant.)

You say you don't have proof-based math courses, but you *will* have Real Analysis after this semester - I'd prioritize getting an A in that, and your math background will be totally fine.  (RA + Linear + Abstract Algebra seems like a heavy load in one semester, so if you have any flexibility, I'd just make sure you can devote the time to do well in RA).

I realize it's already extremely late in this cycle, but you could apply to stats PhD programs right now.  You'd also be a great candidate for top 10 biostatistics PhD programs.  Especially after completing RA and if you're able to start doing research later this year with a Harvard stats prof, I see no reason why you'd need to get a master's before your PhD.  I would just take a year off of school and re-apply.

Hi Bayessays! Thanks for replying! I appreciate your insight. From your experience, do you think my lack of graduate level classes will be a large detriment to my application? That’s one of the reasons I’ve really been considering the masters, especially for more competitive schools. Also, do you think roughly a year of research is enough experience for these applications? Thanks so much for your time!

[bayessays]

If you want to go to a top 10 program, obviously more research and classes can only help.  But your current coursework is fine (stats grad school is not like math where you have to have graduate level courses) and it is not like machine learning/CS grad school (where you have to have publications).

If you still have time to apply this year (obviously not possible with some deadlines), I'd recommend just applying to a wide range of schools in the top 30 or so (and top 10 biostat), and if you aren't happy with the results, getting a master's or getting some more research experience.