Statistics PhD Profile Evaluation Fall 2024

[Medium Point]

Background: International student in the US, male, hispanic.

Institution: Large private institution in Southern California

Degrees: Bachelor in Applied Mathematics, Master's in Applied Math (used to be top ~22 in US news but recently dropped a little).

Undergrad GPA: 3.56/4, Master's GPA: 3.88/4

Awards and Honors: Undergraduate math department award, Dean's Scholarship, School-Affiliated Endowment Fellowship, Dean's list

Teaching Experience: Graduate Teaching Assistant (TA) for calculus 1, calculus 2 for 2 years

Grad Courses: Numerical methods (A), Machine Learning for Data Science (A), Statistical Consulting (A), Applied Probability part 2 (A), ODEs (A-), Predictive Analytics (A, not a math course), Optimization (A), Statistics part 2 (A), Real Analysis (B+, I had an accident that semester which hindered my performance), Statistical Learning Theory (A-, same semester I had the accident).

Research Experience:

• Directed reading under a current PhD student (Fall 2023), worked on generative deep learning (specifically VAE's) and focused on loss functions on bayesian inference models. Created various python simulations and analytical derivations to study the properties of the KL Divergence. Culminated in a 15 minute presentation to the math department.
• Master's Thesis (Spring 2024): Currently working on furthering the above project by trying to develop lower bounds for the KL Divergence when used in Loss functions for statistical learning theory.
• Review paper for a statistical learning class where I worked with my Professor to read a paper on the statistical properties of cross-validation by R. Tibshiriani and T. Hastie and create my own simulations for it. Culminated in a 10 page paper (not published in any journal as it was only a review paper).
• Reading Course (Spring 2022): Supervised by a Professor to read on topological invariants.
• Many Python and R data science projects. The most prominent one was a month long project on energy prices which culminated on a presentation to my class.

Letters of Recommendation: 2 strong (one from the Professor I worked on in the review paper, another from a professor that saw me massively increase from my undergrad to graduate performance and knows of my research projects), 1 semi-strong from a Professor that I did very well in his class (optimization, not technically in the math department but in the engineering department but still a math class - he himself did math as an undergrad).

Universities I'm applying to (all have Jan 1 and later deadlines): My interests are in statistical learning theory/data science/ml. Sometimes it's in the math/stats departments and sometimes in the newer "data science and analytics" departments.

• (Reach): UChicago Data Science PhD, Columbia Stats, Cornell Stats, Northwestern Stats, UWaterloo stats, Oxford, LSE, Imperial College London, ETH Zurich, EPFL
• (More realistic in my opinion): ND Applied Math and Statistics, ND Analytics, Boston University Statistics, Boston University Data Science, Kings College London
• (Supposedly easier): Northeastern Math, Georgetown, Mass Amherst, UVA Data Science, UVA Stats, UMass Amherst

Weaknesses: Undergraduate GPA, LORs in general as I am quite shy/had the accident which limited how much I spoke with my professors although in general I think they know me quite well, No REUs (international so I couldn't do them), no internships (I was taking summer courses as they were covered by my teaching assistantship).

Any thoughts? Please be as realistic and honest as possible. I know my profile is not the worst but also very insecure on it being good enough.

 

Edited December 21, 2023 by Medium Point
typos

[bayessays]

I think most of your reach list is not worth applying to - I don't think you have any chance of being admitted to Columbia or Cornell stats PhDs.  The realistic/easy choices are things I would consider "matches" for your profile but I don't think places like ND/BU/UMass/UVA are safe guaranteed admissions.  Your results will probably be all over the place considering your undergraduate GPA.

[Medium Point]

Do you think they'll weight the undergrad gpa as strongly as the grad gpa?

[bayessays]

Hard to say since you didn't list your specific grades in undergraduate math courses.  If you got Bs in your undergraduate math classes, no, I don't think your graduate GPA makes up for it because usually grad school grades are inflated so 3.88 is average.