Statistics PhD Profile Evaluation/ Advice on what to prioritize

[statsphdapps]

 

Undergrad Institution: Top 5 in U.S (private)

Major(s): Mathematics (Intensive), Statistics and Data Science

GPA: 3.95

Type of Student: domestic white male

GRE General Test:

Taking it in two weeks. Very confident about the Quant section, but not sure I will have enough time to prepare for the writing and verbal

GRE Subject Test in Mathematics:

I took it yesterday. I felt tired and unfocused for the last third of the exam and don't think I did well.

 

Research Experience:

1st Summer: worked in a computational medical research lab

2nd Summer: worked in a computational atmospheric science lab

3rd Summer: worked on a project in a climate science lab that was heavy on statistics and time series analysis

Awards/Honors/Recognitions: Phi Beta Kappa

Pertinent Activities or Jobs: Currently TA'ing introductory Bayesian Statistics

Graduate level courses: 

Functional Analysis (A), Bayesian Statistics (A), Statistical Inference (currently taking) 

Undergrad level courses: 

Math: Real Analysis(A), Abstract Algebra (A), Complex Analysis (A), Proof based Vector Calculus/Linear Algebra I and II (A and A) , Discrete Mathematics (A-), Ordinary Differential Equations (A)

Statistics/ Data Science: Theory of Statistics(A) Introductory Bayesian Statistics (A), Optimization Techniques (currently taking), Data Mining and Machine Learning(currently taking)

Computer Science: Algorithms(A), Data Structures (A)

Physics (I don't know how much Stat programs would care, but some of these were challenging courses): Thermodynamics and Statistical Mechanics (A-), Advanced Classical Mechanics(A), Quantum Mechanics(A), Intensive Introductory Physics I (with lab) and II (A and A)
 

Letters of Recommendation:

-A computational medical researcher with whom I worked with for 3 summers (including the summer after freshman, and two in high school). He writes very strong and generous letters, but I am not sure if this carries much weight being so long ago, especially when the work was more coding and didn't have much statistics. 

- Stats teacher who knows me fairly well, I am a TA in his class now and took 2 courses statistics courses with him ( I am worried I did a mediocre job on his research final project in one class, but he knows me from enough settings to highlight my positive work)

-Math teacher I didn't get to know all that well, but I took two math courses with him, (one being a grad level course), and aced all 4 of his exams. I found his classes and exams very difficult, but I am worried he won't perceive his classes as all that hard

Statistics PhD Programs applying to:

Top Schools: Berkeley, Stanford, UW, Harvard, Chicago, Wharton, CMU, MIT (for operations research instead of stats)

Other schools: To be determined, but I know I should apply to at least 3 or 4 other schools that are easier to get into. Advice on this would be welcomed

Concerns:

-Not doing as well as I could have on the Math subject GRE, but I guess it is too late to retake, and I don't necessarily have to send the scores to most at least

-Lack of any of my letters being able to highlight the ability strength as a statistical or mathematical researcher (rather than just ability to do well on tests, TA, or code at a high level relative to the med students and undergrads in the medical lab). Also maybe it is a red flag that I am not asking more recent research supervisors for letters, but maybe that's justified because I only worked with them for 2 months each. 

-No research publications

-TIME: This is really where I need advice. I only have a month and a half left to take the regular GRE, write a statement of purpose, pick schools to apply to, communicate all the relevant information to my recommenders etc. I am unsure what to prioritize, the regular GRE? communication with letter writers? SOP? should I shrink my list or grow it? Also curious if I have a reasonable shot at top 10 schools and if it is worth the time applying to all of them or if I should focus on applying to schools where I have a better shot.

[abstract_art]

Given you have pretty much a 4.0 from a top 5 undergrad, your math gre is only really gonna matter at Stanford, so I wouldn't worry about it. Also I'd say to not really worry about the GRE, you just need to score above the 90th percentile in the quant section, which I'm guessing isn't going to be hard for someone with your background (most schools don't care about the other sections as long as you don't really bomb them). Lack of letter writers being able to say you can do good research shouldn't be much of a problem since most applicants don't have a ton of research in general. Also as you can see from the stickied post in this subforum, the sop doesn't really matter a whole lot.

 

You have a good shot at pretty much any stats program, so I'd say apply to where you want to go, and if you're really worried about getting shut out everywhere throw in a few larger programs (like Wisconsin, Michigan, Texas A&M, NC State, Penn State, Minnesota, Purdue, etc.). Since you don't have a whole lot of time left, just prioritize making sure you finish all of your apps on time, and making sure all of your letter writers know when they have to submit letters by.

[lemma]

I have a strong, strong suspicion we went to the same undergrad. If so, try as hard as you can on your applications, but relax. People with GPAs in the 3.7+ range routinely get into top schools, and I have friends with much weaker backgrounds who got into solid programs. There are exceptions, but they are the exceptions rather than the rule. 

Anyway, enjoy senior year while you can. You only get to do it once. And if we did go to the same undergrad - I loved thermodynamics and statistical mechanics, but the top stream of classical mechanics was even more fun

[statsphdapps]

Thank you so much marmle and lemma for these thoughtful responses and the advice on what to prioritize. It has helped me feel much more confident and eager to dive in and start filling out the applications and getting in touch with my letter writers.

Lemma -Classical mechanics was easily my favorite physics class followed by thermodynamics and statistical mechanics! Given that these were also your favorites, it sounds extremely likely we went to the same undergrad. Perhaps we have even been in multiple classes together if you are senior this year too  

 

[lemma]

I was a senior when you were a freshman, so sadly we probably didn't overlap, but there's a good chance we've been taught by the same professors! I miss that place. 

As I said, most people I know who scored a 3.7+ who wanted to a PhD got into top 5 schools (thinking mathematical sciences because that's what I know best). I have three friends across three years who didn't, but their grades were at the lower end of that spectrum and two of them were non citizens. There are never guarantees, but it sounds like you've pretty much done the best job of what you can control.

[statsphdapps]

Update: I wound up scoring and 820 on the Math Subject GRE (82 percentile)! Given my strong GPA and math courses from a well-known undergrad institution, would it help or hurt me to send such a score to schools that do not require the subject GRE but strongly recommend it such as Chicago and Washington (I feel like I'm in a grey area as to whether or not sending the score will be helpful). What about for schools that say the test is optional, and you can report scores but they don't strongly recommend it?)

 

[yenhcz1995]

On 11/24/2017 at 9:54 AM, statsphdapps said:

Update: I wound up scoring and 820 on the Math Subject GRE (82 percentile)! Given my strong GPA and math courses from a well-known undergrad institution, would it help or hurt me to send such a score to schools that do not require the subject GRE but strongly recommend it such as Chicago and Washington (I feel like I'm in a grey area as to whether or not sending the score will be helpful). What about for schools that say the test is optional, and you can report scores but they don't strongly recommend it?)

 

I got the same score and would also like to know the answer to this question.