Stat PhD 2021F

[dlsxyt]

I graduated this June, currently working as a software developer at a top 5 bank. I'm interested in doing research in bayesian statistics / nonparametric statistics. Something bad happened in my first year so I got only ~3.8 in first two years of uni. 

Undergrad Institution: Canada top 3
Major(s): Statistics
Minor(s): Computer Science
GPA: 3.89(last two years 3.97)

Type of Student: Asian Female

GRE General Test: 
Q:165
V: 161
W: 4.0

 

Applying to: Statistics PhD

 

Research Experience: One year at a relatively large hospital, worked mainly with logistic regression, co-authored an abstract. Other than that I do not really have much research experience.  I have several bayesian course based projects.

Letters of Recommendation:  Probably a decent one from summer reading on math. One from AP teaching ML(got 99). I am hesitating between a prof who Ive taken course with, or my hospital research prof. Neither of them knows me well, but the former is a COPSS winner while the latter is from the only research I've done.

 

Math/Statistics Grades:  

1st year: Real Analysis(B+), Linear Algebra I(A+), Linear Algebra II(A)

2nd Year: Advanced Cal(A-), and probability I(A+), Prob II (A-), 

3rd Year: Stochastic Process(A+),  Spatial Stat(A), ML I(A), Applied Stat(A+)
4th Year: Stat Computation(A+), Multivar Data(A), Mathematical Stat I(A), Mathematical Stat II(A+), ML II(A+),

And a bunch of CS courses: Neural Network, Database, Algorithm, etc, at least A
Planning on Applying to: 

Duke, UCLA, UW, UPENN, UNC Chapel Hill

 

What are my chances of getting into some of these schools, and what are some other schools I should look at? Thanks for your help!

 

Edited August 21, 2020 by dlsxyt

[bayessays]

I don't generally comment on international admissions, but clearly your profile is very strong and I think you definitely have a good shot at UNC and UCLA. Penn might be a stretch.  If you are interested in Bayesian statistics and nonparametrics, UT Austin is the place to go for Bayesian nonparametrics. 

[dlsxyt]

Thank you so much. You really gave me some confidence.

[Casorati]

You have very strong grades coming from a school that is known for grade deflation. I think you have enough math courses but some additional courses such as measure theory/functional analysis may strengthen your profile, considering that many international applicants have very rigorous training in mathematics. Your list looks good and I can see you getting into a few of them. I would apply to 10+ schools with some safer options at the level of Penn State/UC Davis/Minnesota since admissions are competitive for international students. I have a friend from your school with similar background as you ended up into one of the above schools. I would also apply to your undergrad institution. If you are interested in Bayesian statistics and non-parametrics, UBC also has several people working in this field.

Edited August 21, 2020 by Casorati

[insert_name_here]

Unless you had a whole bunch of deep conversations with the COPSS winner, go with the hospital person you did research with for your letter.

I'd add a school or two to your list, but it looks reasonable (Maybe a "dream" school, and one or two lower ones). You've got great grades from a great school, don't stress.

[dlsxyt]

11 hours ago, insert_name_here said:

Unless you had a whole bunch of deep conversations with the COPSS winner, go with the hospital person you did research with for your letter.

I'd add a school or two to your list, but it looks reasonable (Maybe a "dream" school, and one or two lower ones). You've got great grades from a great school, don't stress.

Hello, thank you for your suggestion. May I ask for more detail? I didn't have a deep conversation with neither tbh. I understand the COPSS winner might just gave me a general one, but the one from the hospital(Not a statistics prof) won't be very outstanding. I am very hesitated. Thank you!

Edited August 22, 2020 by dlsxyt

[insert_name_here]

The main purpose of letters is to demonstrate your research potential. So, it's a good idea to get letters from people you did research with. If you co-authored something with them, surely you spent time with the hospital prof, or one of their students/postdocs. Particularly given you have no other research letters.