Jump to content
nullhypothesis

Statistics PhD 2020 Applicant Profile - No real analysis

Recommended Posts

Posted (edited)

I very recently decided that I wanna get a PhD in stats - because of that, I have some fundamental lack of courses that I feel are required by most stats phd programs. Just wondering which schools I should target and what my chances look like?

Undergraduate Institution: Top 20 State school

Major: Math and CS (Joint)

GPA: 3.80

Statistics/ML Courses: Mathematical Statistics I/II (A), Probability (no measure theory) (A), Stochastic Processes I/II (A),  Machine learning (Graduate) (A), Data Science (A), Pattern recognition (A)

Math Courses: Calc I-III (A), Linear Algebra (lower division) (A+),  ODE (lower division) (A+), Abstract Algebra I/II/III (A+), Convex optimization (graduate) (B+), Combinatorics (B+), Intro to proofs class (A), Numerical Analysis (A+) 

MISC: Data Structure and Algorithms (A), Advanced algorithms (A), Theory of computation (A), Recommender systems and data mining (A+), Game theory (A-)

Note: I'm planning on taking a real analysis course from UIUC via https://netmath.illinois.edu before applying. 

Another fact I thought might be worth mentioning - I'm graduating in 3 years.

GRE: 168; 164

Awards: Phi Beta Kappa, other misc. Academic / Research awards (2)

Research Experience: Worked on mathematical neuroscience research and currently two papers (second author) under review - not much stats involved at all. 

Recs: Very strong from one math/cs professor and from my research PI. 

Work experience: Data science / ML internships at two fortune 100 companies. Starting full time job at one after college. 

 

I was planning on applying to: 

Duke University

Stanford University

Harvard University

Carnegie Mellon University

Columbia University

Princeton University 

University of Chicago 

John Hopkins University

New York University

Penn State

Notre Dame

 

Thanks!

 

Edited by nullhypothesis

Share this post


Link to post
Share on other sites

I don't understand - you are taking analysis before applying, so there's no issue. You have a good math background anyways, so it isn't the biggest deal, but schools like Stanford are going to be full of people who took more math.  I'm going to assume you're a domestic applicant?

 

You need to do way more research on programs.  Stanford, Chicago, Harvard, Columbia are going to be huge reaches - some of these schools require math subject test. Duke, CMU are super competitive and are probably where your reaches should be lying if you want to apply to top schools.  NYU accepts like 1 student a year, so that's probably out.  Princeton doesn't even have a statistics PhD. Then there are ND and JHU, which don't really have dedicated departments and probably wouldn't be called top 50 programs. 

 

So basically, you're shooting yourself in the foot by applying to mostly schools that are reaches, and then 4 schools that barely have departments.

 

You should be applying to way more schools surrounding the Penn State level, in my opinion.  You went to a state school - where are all the state schools? Michigan, Ohio State, Minnesota, Illinois, UCLA, Wisconsin.  You could get into some of these programs today probably and have a very successful career from them.

Share this post


Link to post
Share on other sites

I agree with Bayessays.  I think I should add that one mistake people sometimes make in choosing PhD programs is choosing a bunch of prestigious private schools on name reputation, without giving proper attention to public programs that are often larger (meaning, relatively easier to get into) and potentially better.  I know I made that mistake on a first pass, and was fortunate to learn otherwise here on gradcafe.  Getting a PhD at NC State or Illinois (etc.) might not sound as prestigious to an uniformed stranger, but academics or smart industry people will know that those are strong programs.  Just in case you haven't looked at it yet, the USNews grad school rankings in statistics are a good place to start, and since those are kinda confusing in format, feel free to contact me with questions via DM.  They aren't everything, though, so don't stop there!

Share this post


Link to post
Share on other sites
Posted (edited)

Agreed with bayessays and Geococcyx. Stanford, Chicago, Harvard, Columbia, Penn, etc. are very difficult to get into and have a strong preference for applicants who have taken a lot of advanced mathematics -- many of the students they accept will have already taken the Masters-level mathematical statistics sequence (based on Casella & Berger), as well as advanced math classes like graduate measure theory or graduate-level probability theory. I think the best range of schools for your profile would probably be Texas A&M to Purdue, and you should have a very good shot at schools like UIllinois and Rutgers. If you are interested in academia, I note that there are PhD graduates from these aforementioned institutions who end up at postdocs at places like Harvard, Penn, Columbia, Duke, etc. So if you work with a good PhD advisor and excel at one of those programs, you should be able to land a good postdoc.

Edited by Stat PhD Now Postdoc

Share this post


Link to post
Share on other sites

Thank you everyone, really appreciate the input, here's my revised list: 

Duke University

Columbia University

University of Chicago 

Cornell University

Carnegie Mellon University 

Penn State

UCLA

University of Michigan 

Texas A&M University

UNC Chapel Hill

 

What do you guys think about this one?

Edited by nullhypothesis

Share this post


Link to post
Share on other sites

You're about half way there. You still have to add lower schools - none of those is even close to a safety or even a safe target (Edit: I do think you're likely to be accepted to TAMU).  I'd take off Chicago and Columbia and add 5 schools that are below every school on your list. 

Edited by bayessays

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×

Important Information

By using this site, you agree to our Terms of Use and Privacy Policy.