Stats PhD (Profile and School Recommendations)

[WarAndPeace]

Hi Everyone,

Any advice on schools to add/remove given my background and research interests?

 

Undergrad Institution: Large Public School, around top 100 (US news)

Undergrad Major: BA in Mathematics

GPA (Undergrad): 3.72, overall and major

Type of Student: Domestic

Relevant Courses (Undergraduate): 

Calc I/II/II (A-, B, A-), Analysis I/II (A-, A), Linear Algebra (A), Discrete Math (B+), Differential Equations (A), Applied Probability Theory (A), Algorithms (B+), Applied Probability Theory (A), Math Stats (B-), Machine Learning (A), Econometrics (A-)

Research: 

• 1 semester of math research in number theory, internship in machine learning
• 1.5 years as a research associate out of undergrad at a top business school (Wharton/Columbia/Stanford). Some applied statistics, nothing real theoretical. 

_____________________

Graduate Institution: Public School, around top 30 (US News)

Major: MS in Mathematics

GPA (Graduate): 4.0

Relevant Courses (Graduate): Real Analysis I/II (A, A), Complex Analysis (A), Probability I/II (A, A), General Linear Models (A), Statistical Estimation (A), Multivariate Stats (A), Statistical Learning (A), PDEs I (A), Hypothesis Testing (Fall), Stochastic Calculus I (Fall), Algebra I (Fall)

Research: 

• Top Artificial Intelligence conference involving bayesian inference
• Journal Publication, Finance and Economics

Recs: 3 strong letters: 1 from probability grad class, 2 from research

GRE General: Q/V/A: 168/163/5

GRE Math Subject: Planned on taking in September, but not sure whether it will be offered in the fall with the COVID19 situation.

_____________________

Research Interests: Bayesian inference, High dimensional inference, Graphical models, Stochastic analysis (stochastic differential equations)

Prospective Schools: Statistics only, but I am also planing on applying to several Applied Math/CS PhDs which cover the research interests above.

• US: Berkeley, Duke, NCSU, Columbia, Yale, Michigan, Texas
• Non-US: Toronto, Oxford

 

 

 

[bayessays]

I could see you getting into some of those schools, but I would still consider most of not all of them to be in the reach category. I would add schools at the OSU/UIUC level. 

[WarAndPeace]

3 hours ago, bayessays said:

I could see you getting into some of those schools, but I would still consider most of not all of them to be in the reach category. I would add schools at the OSU/UIUC level. 

Thank you, I will add some schools in that tier to my list. Do you have any advice on maximizing my chances, or improving my profile? For example

1. More pure math (delay application a year). In the spring I could take some combination: Algebra I, Algebraic Topology, Differential Geometry, Functional Analysis, Harmonic Analysis.
2. Scoring high on math subject test.
3. Theoretical statistics research (hardest to gain out of the 3 though).

Edited May 10, 2020 by WarAndPeace
details

[bayessays]

I think you have plenty of math.  The subject test may help for some schools, but I'm not sure whether schools that don't recommend it even look at it.

A lot will probably come down to your research letters.  Are you first author on these papers and/or significantly contributed and will well-known letter writers write great things about you? It's hard to tell these things.  If these are stellar publications and recs, I could see you getting into quite a few places on the list.

But admissions are also getting harder and I had a similar record to you in a lot of ways (some shaky undergrad grades, good MS grades, publications) and I only got into 1 of the schools you listed out of 5.  I suspect you may be able to do a little better, but I don't think the list of schools you gave has enough safe options for all but the most qualified applicants. 

[WarAndPeace]

4 minutes ago, bayessays said:

I think you have plenty of math.  The subject test may help for some schools, but I'm not sure whether schools that don't recommend it even look at it.

A lot will probably come down to your research letters.  Are you first author on these papers and/or significantly contributed and will well-known letter writers write great things about you? It's hard to tell these things.  If these are stellar publications and recs, I could see you getting into quite a few places on the list.

But admissions are also getting harder and I had a similar record to you in a lot of ways (some shaky undergrad grades, good MS grades, publications) and I only got into 1 of the schools you listed out of 5.  I suspect you may be able to do a little better, but I don't think the list of schools you gave has enough safe options for all but the most qualified applicants. 

This is very helpful. I appreciate the insight.

I'm not first author, but significantly contributed (e.g. ~2 years of work on the journal, some theory and all implementation on the conference). My letter writers on the research side are well-known and at an IVY institution, but not in the statistics department: 1 in computer science (heavy emphasis on bayesian methods), and 1 who is a pioneer in applied mathematics (MCMC and the likes). I've known/worked with them for about 3 years, so I expect them to be stellar. 

I gather from your reply that it will be a crapshoot, more or less, for the more difficult schools as this point. Always difficult to accept! Thanks again. 

[captivatingCA]

I agree with @bayessays. I think my letters are what pushed me over the edge. It sounds like your letters will be great, so I think you could have a good shot at the schools you mentioned. If you haven't already, I would show your recommenders your list or ask them what schools/departments they think you're competitive for. That definitely guided my application decisions.

I also wanted to add that even though your list has a decent range in terms of ranking, the departments you mentioned are all pretty selective either due to size or prestige or both (e.g. Yale). I think you would benefit from applying to a few more big departments. Texas A&M comes to mind since they have a significant Bayesian presence. Additionally, it may be a good idea to focus your applications to one field in particular (if you aren't already).