EntrepreneurIll4719

Hello, hope you are doing well. I'm applying to PhD programs in statistics, biostatistics and related fields for Fall 2023 entry. I would really appreciate some thoughts on my profile and potential destinations ahead of this admissions cycle. I am broadly interested in the careful and responsible application of statistical learning, especially to solve problems in medicine. Ideally, I would like to help researchers in biomedical fields by developing methods to deal with challenges and complexities that they may come across in their data, such as high dimensions, dataset shift, imbalance, etc., in a way that is as reasonable and interpretable as possible. Applications in other fields are not completely out of the question, but applications in medicine is what excites me the most. Type of student: International Asian Male Undergraduate Institution: St. Olaf College Undergraduate Major: BA in Mathematics, with concentration in Statistics and Data Science GPA: 3.80 Relevant Coursework: Calculus II (A), Linear Algebra (A), Differential Equations I (A), Real Analysis (A), Statistics for Science (A), Calculus III (A-), Abstract Algebra (A), Introduction to Data Science (A), Introduction to Machine Learning (A), Probability Theory (A-), Statistical Modeling (A-), Differential Equations II (A), Combinatorics (B+) Graduate Institution: Wake Forest University Program: MS in Statistics GPA: 3.82 (after including Fall 2022 grades) Relevant Coursework: Linear Models (A-) Probability Theory (A) (Took a similar course early in undergrad) Real Analysis (A) (Took a similar course early in undergrad) Stochastic Processes (A) Causal Inference (A-) Advanced Statistical Inference (B) (This class had some internal issues; one of my recommenders will explain my poor grade) Generalized Linear Models (A) (midterm grade) Multivariate Statistics (A) (midterm grade) Time Series and Forecasting (A) (midterm grade) GRE: General Test: 165V/165Q. Have not taken the subject test. Not sure if I should submit this since most programs are GRE optional. Some even recommend against submitting at the moment. Research Experience: Currently working on a master's thesis. We investigate the application of active (machine) learning strategies to build generalizable classifiers efficiently, and study the impact of performing inference when using the predictions of these classifiers instead of ground-truth labels. The problem is motivated by data from cell biology where there is a significant class imbalance, an abundance of unlabeled data and multiple conditions to generalize across. There is potential for a paper to come out of this but definitely not in time for the admissions cycle. Work Experience: Worked with a health insurance company for about 5 months to create a recommendation engine that matches customers with the most suitable insurance plan for them. Various machine learning techniques were used to devise the method, an app was the final product. Letters of Recommendation: One letter from my current research advisor and statistics professor who has also had me in class (GLM, likely with an A). One letter from a senior statistics professor who knows me very well, has had an openly high opinion of me and has had me for 4 classes (all A's). One letter from my undergrad academic advisor who oversaw a practicum project and had me in a class (PDE, with an A). I expect all of them to be strong letters of recommendation. Program Shortlist: This is simply a tentative shortlist based on my research. I will apply to around 10 colleges that fit me best. I fear this list may be top heavy; nevertheless, I was encouraged by my advisors to include them as they believe the programs are great fits for me. I would appreciate some "matches" and "safeties" (although I understand that it may not quite work that way for PhDs). I like all of the ones I have listed and will be genuinely happy at any one of them, hopefully I can discover more with your help. Most of the programs I will apply to are "Statistics PhD"s, due to their flexibility as well as their interdisciplinary and computational focus. You may notice that most of these institutions are situated in and around the east coast, particularly in New England and North Carolina. This is because I have friends and family here. I realize that there are excellent programs everywhere, and some of them may be great fits for me as well, but all else equal I would like to prioritize these areas. Unfortunately, the good programs here tend to be very competitive admissions-wise, so if you have alternatives elsewhere I would absolutely consider it. Statistics: In statistics PhD programs, I would like to see a focus on applied, computational statistics, and interdisciplinary work, where collaborations with schools of computer science, medicine and engineering are easy and frequent. I am less interested in finance, economics, or social science applications, although anything interdisciplinary and applied appeals to me more than mathematical and statistical theory. Carnegie Mellon University - Statistics PhD (obvious fit due to machine learning focus, thesis advisor went here) Johns Hopkins University - Applied Math and Statistics PhD (working with the Institute of Computational Medicine sounds splendid) Cornell University - Statistics PhD (they seem to be strongly connected to the Department of Computational Biology and encourage interdisciplinary work) UNC Chapel Hill - Statistics and Operations Research PhD (really like the emphasis on application, and appreciate the strength in stochastic modeling as well) NC State - Statistics PhD (bioinformatics and statistical genetics group is especially interesting) University of Pittsburgh - Statistics PhD (fan of the collaborations with the school of medicine, cancer institute, etc.) Boston University - Statistics PhD (Lots of interdisciplinary research, plenty in biomedical areas such as bioinformatics and neuroscience as well) Yale University - Statistics PhD (there seems to be an increased focus on data science, affiliation with Yale Medicine and Public Health is a plus) Biostatistics: Although I initially thought that "biostatistics" PhD programs would be an obvious fit, the flavor of all of the biostatistics areas do not appear to be the same. I am not particularly interested in many classical biostatistics areas such as clinic trial design, causal inference, epidemiology, ecological and population-related statistics, etc., and am more interested in computational areas such as bioinformatics. Of course, there will always be overlap, but roughly, I am more interested in the statistics that is more adjacent to "machine learning in biomedical research" than to public health areas. I am also worried about NIH funding and how it might impact my application due to my international status. University of Pennsylvania - Biostatistics PhD (situated within the school of medicine, really interested in the intersections with other areas of the Biomedical Graduate Studies umbrella). Duke University - Biostatisics PhD (situated within the school of medicine, Duke StatGen and GCB seem like very cool collaboration areas) Computational Biology: There's a slew of very interesting programs in this intersection. However, most of them seem completely unavailable to me, because I have absolutely no formal biology background. While this usually doesn't seem to be a problem for "biostatistics" programs, some biology coursework seems strongly encouraged by these programs, and looking at successful applicants, they typically have at least some biology background, if not explicitly bioinformatics/biomedical engineering focus. How I wish I just took a couple of biology courses instead of rushing through undergrad in just 3 years... I included a couple, in case anyone had any thoughts on them. I might cross-post this part to the biology forum to get their thoughts as well. CMU-Pitt Joint Initiative - Computational Biology PhD (bringing together the strengths of the two institutions, probably too competitive for me but I had to include it because it sounds very cool) Dartmouth College - Quantitative Biomedical Sciences (I admittedly don't know much about Dartmouth in statistics or medicine, however it was the only quantitative biology PhD that didn't explicitly require a biology background so I included it) Reflections: As I mentioned, the list is probably top heavy and has pretty much neglected programs outside of New England and North Carolina. I would really appreciate thoughts on both my competitiveness for the programs I've listed as well as some new ones, wherever they may, that might be good fits for my profile. Thank you so much for your help!

Hello, hope you are doing well. After looking at several profiles, I'm coming to the realization that my application is quite weak and I'm looking for the strongest schools I can get into. I graduated on May 2020, and I am currently waiting for OPT approval. I am applying for Fall 2021. Absolutely any help is greatly appreciated. Type of student: International Asian Male (Indian) Undergraduate Institution: Small mid-western liberal arts college (St. Olaf College) Undergraduate Major: Mathematics, with concentration in Statistics and Data Science GPA: 3.8 Overall Relevant Math and Stats Courses: Calculus II (A), Linear Algebra (A), Differential Equations I (A), Real Analysis (A), Statistics for Science (A), Calculus III (A-), Abstract Algebra (A), Introduction to Data Science (A), Introduction to Machine Learning (A), Probability Theory (A-), Statistical Modeling (A-), Differential Equations II (A), Combinatorics (B+) GRE: Haven't taken yet. In the one practice test I took, I scored 170 Quant, 166 Verbal. Research Experience: None. Work Experience: Worked with a health insurance company to create a recommendation engine that matches customers with the most suitable insurance plan for them. Various machine learning techniques were used to devise the method, an app was the final product. Was involved in every stage of the project. The whole thing was about 5 months long. This is my only relevant work experience. Hopefully I can get some more experience in the time before I go back to grad school. Not that it will help me with my application, would be nice nevertheless. Letters of Recommendation: Two from math professors who view me favorably. One of them was also my academic advisor. The final letter would be from my supervisor from my work experience. They should all be strong letters as they knew me well, but probably not phenomenal. Colleges Shortlist: Please take this with a grain of salt. I have no idea how selective they are and odds are I am grossly overestimating my chances. Although it would be nice to keep a couple of them, I would really appreciate some new suggestions if you think they are all reach. University of Minnesota (Twin Cities), University of Michigan (Ann Arbor), University of Washington (Seattle), University of Wisconsin (Madison), University of North Carolina (Chapel Hill), University of California (Los Angeles), University of Illinois (Urbana-Champaign), University of Chicago, Duke University, Georgia Tech, University of Connecticut. A few more things: Do you think it would do well for me to, once I get my OPT, use it as much as possible to strengthen my resume and work for a year or so more, and then apply to graduate programs? Do you think I would have a better shot at, say, the top 25 or so universities in my field? Do you think that (ranking) matters, is there a big difference between top 10, top 25, top 50 etc? I have absolutely no research experience and I have no idea how I'm supposed to get it. Could you help me by letting me know where I can find opportunities for post-undergraduates in my field (statistics/data science)? Thank you for your help, I really appreciate it!