Stats PhD Fall 2021

[physics2stats]

Hi All,

I'm a pretty non-traditional applicant to Stats PhD programs -- I have a few questions, and am interested to see what y'all think of my chances!

Student Type: Domestic White Male

Undergrad: Top 10 overall (Private) (US News Rankings)

• Major: Physics (Chemistry minor)
• GPA: 3.88
• Math Coursework: Calc 2-3 (A), Linear Algebra (A), Diff Eq (B), Complex Analysis (A), Numerical Analysis (A), Geometry Seminar (A)
• CS Coursework: Intro Programming (A-), Computer Architecture (A+ -- taken post-grad to prep for CS grad school)
• Other: Upper-level undergrad physics: Quantum I (A), Classical Mechanics (A-), Thermal Physics (A), Special/General Relativity (B+), Nonlinear Dynamics (A); Intro and Organic Chemistry Sequence (all A's)

Graduate: Top 15 Computer Science MS (Public) (actually entering second year in the PhD program, but have decided I'm more interested in stats)

• Major: Computer Science
• GPA: 4.0
• Coursework: Intro Artificial Intelligence (A), Intro to Theory of Computing (A), Intro to Algorithms (A), Probability Theory (A), Analysis I (A), Stochastic Processes (A), Linear Optimization (A), Machine Learning (Fall 2020), Nonlinear Optimization I (Fall 2020), Undergrad Math Stats (Fall 2020)

GRE: Took in 2013, so it's expired and I'll have to retake. Scores then were V: 164, Q: 170, A: 5.0

GRE Math Subject: N/A

Research:

• Publications: First author for a journal article on biological optics from undergrad supervised by a biology prof (cited 14 times as of posting this), Summer research on reinforcement learning under a physics postdoc (paper posted on Arxiv -- cited 2 times), worked on uncertainty quantification for ML methods in computational materials science since January 2020 -- hoping to get a paper by the time I submit applications this winter
• Misc: I'll be doing some data analysis for an education research group this fall -- (trying to apply ML methods to say something meaningful about user data for online education games)

Letters of Recommendation:

• Planning to ask my three research supervisors (1 biology professor, 1 research scientist at Facebook now, 1 materials engineering professor). 2nd and 3rd should be strong. I think 1st should be strong too, but it's been a while since undergrad... I might ask my supervisor for my education research this fall if it goes well?
• Should I also ask a math professor? I don't think any of them would have a ton to say, other than "he did well in my class," which I'm thinking already comes across from my grades?

Research Interests: Roughly: Developing applied/computational statistics methods for problems in the natural and social sciences

Miscellaneous: I took 5 years between undergrad and my current grad program. In that time, I taught high school math and physics and got a MA in Teaching (3 years), and then worked for an education technology company designing online K-12 science/math curriculum (2 years). I really enjoyed both of those, but decided to go back to school because 1) I didn't see myself being satisfied in those career paths for an entire career, 2) I missed being in school and learning cool stuff all the time, and 3) especially after the reinforcement-learning research experience, I definitely bought into the "big data" hype, and wanted to be a part of it. I managed to sneak into a solid CS program, but once here I quickly gravitated toward the theory/math/stats side of things, and started considering mastering out and looking into stats. (As you can tell from my undergrad coursework, getting into top stats programs would have been tough without the probability/stats/analysis background that I've gotten in my current grad program.)

Schools:

• Duke, UNC, NC State, Wisconsin, Michigan, Harvard, Texas A&M

Questions:

• Thoughts on getting rec letters only from scientists in Bio, Physics, and Engineering?
• I'm aware my background is a pretty non-traditional -- I feel like it's given me a chance to grow personally and professionally, but any takes on how an admissions committee might view it?
• How much explanation do I need about transferring from CS PhD program to stats? Is just saying something like "I decided I want the rigorous theoretical foundation I would get through a Statistics PhD" sufficient? (Or, can I just frame it as an MS program I'll finish in May 2021 and leave it at that?)
• General thoughts on my chances at the schools I listed? Or suggestions for others?

Thanks!

Edited August 25, 2020 by physics2stats
typo

[Stat Assistant Professor]

1. Recommendation letters from those professors should be fine. Maybe ask them to emphasize that you're also a strong math student who has taken some advanced math like Analysis I, complex analysis, and optimization. 

2. I think admissions committees will view someone with a physics BS from a top 10 undergrad and a ~3.9 GPA very favorably actually.

3. In your case, I don't think you need to say much about transferring PhD programs. It would be different if you were transferring from one Statistics PhD program to another, or if you had already completed 3 years in a PhD program. Since you completed only one year, you can just say, "I will be completing my Masters in computer science in May 2021 and am looking to further my education in Statistics" and then talk a little bit about what research you could potentially see yourself doing. 

4. Of the schools on your list, I would say your chances at Duke, UNC, NC State, Wisconsin, UM and TAMU are all very good for your profile, since you have the math background, a degree in physics (and soon degree in MS), and a strong pedigree. I would say your chances are well above average at all those schools. Harvard is a possibility too, but they (like many of the Ivies and UChicago) seem to admit very few domestic students, so it's not a sure thing. But you should definitely apply there.

[physics2stats]

Thanks for the quick response -- that's super helpful and comforting!