Stat Assistant Professor

After you updated your post with your info, I would have to agree with the posters above that your profile is very strong. You could probably apply to all top 10 programs, and I'm sure you would get into at least a few of them. UC Berkeley and UW definitely seem plausible, as does Stanford if you can score well on the Math Subject GRE. If you are more flexible about your geographical preferences, you could probably get into really good schools on the east coast or midwest as well.

- If you have not already taken it yet, take real analysis. And prioritize taking math classes rather than undergrad stat courses. Get A's in your math clases. - Score well on the math section of the general GRE - ideally 164+. Unless you are planning to apply to Stanford or a school that "strongly recommends it," you don't need to worry about the Subject GRE. - It seems like research experience is becoming more common for applicants to Stat PhD programs, so try to get some if you can.

I'd recommend taking more math in the fall semester of your senior year, apart from Analysis (maybe "Advanced Linear Algebra", i.e. linear algebra with proofs, and one other upper division math class). If you do well in these classes, you should stand a decent shot at most of the UC schools and possibly UW and UC Berkeley. I think your profile is very solid, and the research experience is a plus. But you should take more math.

If the Canadian programs you listed as unfunded, then I imagine you should be able to get into at least one of them. If they are fully funded even for MS students, then they might be more competitive.

You stand a good chance getting into a good Masters program. I think UIUC, UCLA, and TAMU sound perfectly reasonable. I would avoid any "Applied Statistics" MS programs. I think it is better (for both industry and possible preparation for future PhD programs, if you're contemplating that) to make sure you attend an MS program that requires a year-long sequence in mathematical statistics (usually based on Casella & Berger) and a class on Theory of Linear Models.

I'm sure you could get into some Biostat MS program. I would consider retaking the GRE to try to get a slightly higher score on the Quantitative section (you can take the test from home now, according to ETS -- no need to go to a test center).

Oh oops, I somehow missed that the OP is applying to Statistics PhD programs rather than math PhD programs. In that case, disregard what I had written earlier. In that case, I would just review some Calculus, linear algebra, and maybe upper division undergrad-level probability and statistics and undergrad-level real analysis (these latter topics aren't as essential as having finesse with Calculus or LA, but basic familiarity with them might make the Casella & Berger mathematical statistics sequence a bit easier).

Some math PhD programs like UCLA and UC Berkeley require incoming PhD students to take a "preliminary exam" or "basic exam" upon arrival that covers upper division undergrad material like proof-based linear algebra, real analysis (e.g.at the level of Rudin's "Principles of Mathematical Analysis"), complex analysis, abstract algebra, etc. You could spend some time practicing the publicly released prelim exams from different departmental webpages to make sure you will be able to pass these exams. Other (usually mid-ranked or lower-ranked) math PhD programs basically re-teach you undergrad abstract algebra and analysis in the first year of the program before covering measure theoretic analysis and graduate-level abstract algebra and Galois theory. In that case, I think it is still worth it to review linear algebra, real analysis, complex analysis, and abstract algebra and practice the prelim exams from departmental webpages -- that way, you will be very well prepared for the first year of the PhD program, and that always makes the transition from undergrad to grad school a bit smoother. This is what I did the summer before I began my PhD (though I'm in Statistics). Although I had previously taken Casella & Berger Mathematical Statistics, applied regression, etc. it had been almost two years since I finished a Masters, so I was a bit rusty. So I did every single past first year exam that was released on my department's website, and that helped a lot. I also reviewed real analysis and linear algebra before beginning.

Your plan of action sounds great. You could apply to some larger departments that have biostatistics and statistics together under the same umbrella (e.g. NCSU, UWisconsin, and FSU Statistics all house biostat in the Statistics department and allow PhD students to do a Biostatistics concentration for their PhD). Many Statistics departments -- even ones that are known to be more theoretical,-- will also have a few faculty working mainly on applied problems, including those motivated by biology/genetics (e.g. Stanford has Susan Holmes, UPenn Wharton has Shane Jensen and Nancy Zhang, etc.).

Yes, the publication should enhance your application. Not necessarily because it is very relevant to biostatistics but because it is a clear sign of "research potential." I would ask your research supervisor for a letter of recommendation. That letter, plus one recommendation letter from a professor who can state how adept you are at advanced mathematics, should make your overall application very strong. I don't think your freshman years should hurt your application much since you got all A's after that in upper division classes. Best of luck.

You should be fine. There are a lot of Statistics and Biostatistics PhD students who never took a single statistics class before enrolling in their PhD program.

I don't think UW or Michigan is a reach for your profile. As long as you explain the medical issue your freshman year in your application (or better, have an LOR writer explain it in their letter), it shouldn't be a huge issue for many good programs. Your GPA is pretty good, you went to a top school, and you have A's in advanced classes like Real Analysis and Galois Theory. The research experience in experimental physics is also a plus. For Statistics, I think you should actually aim for schools like Duke, Wisconsin, NCSU, UNC, and possibly UC Berkeley. The Ivies, Stanford, MIT, Caltech, etc. are all really difficult to get into, but you could try a few of them too if you wanted. MSU and UCI are both fine programs with good academic placements (I saw that UCLA Biostatistics just hired someone who got his PhD from UCI, and there are other alumni from these programs at places like UF and ASU) -- but I think you could definitely get into a top 20 Statistics program if you applied to some.

I think you can likely get into a Masters program provided that you have good letters of recommendation and explain your undergrad academic performance/emphasize any upward trend. Although your GPA isn't great, it is above the minimum threshold not to be auto-rejected. Masters admissions are also usually more lenient, since they are rarely funded. Finally, UT Austin is a well-respected school, so you have that going in your favor. I have also seen some individuals use Masters programs as "stepping stones" to get into PhD programs. These folks didn't have the best undergrad grades, but they did the Masters to show that they could get A's. Then they went on to do a PhD program (in math/applied math/stats). If you wanted to do that, you probably could -- though you would likely have to temper your expectations for PhD admissions, in terms of what tier of programs you could get admitted to... but mid/lower-ranked schools might take a chance on you. If your goal is just to get a Masters and then work in industry, I could see you getting into some MS program.

The top schools on the USNWR list look about right to me (NYU Courant, UCLA, and MIT). I would suggest you look at more than only rankings. Make sure there are actually enough faculty who are: a) doing research in your area of interest, and b) who have solid placements for their advisees. Are they placing their former students into top postdocs? If the school is a top-tier one like NYU, UCLA, MIT, Princeton, etc., the answer to part (b) is assuredly "yes." But beyond that, I would make sure there's a big enough group of researchers there. Some programs might be stronger in some area of applied math like PDEs/fluid dynamics or numerical analysis/optimization than others.

Did you have any meaningful interaction with your professors in your online classes? You could ask one of the professors in your remote advanced math classes for a letter of recommendation. In their letter, they can describe that the classes you took were equally as difficult as the in-class ones and also involved rigorous proofs and derivations. If you interacted with these professors in any meaningful way, then I think it should probably be fine. You can also explain in your SOP that because you were working full-time, the online courses made the most sense for your schedule. In fact, stressing that you worked full-time *and* took these additional classes because you were so motivated to pursue a PhD in Biostatistics may help your application. For comparison, I know one PhD graduate from Harvard Biostatistics (now a postdoc at Princeton) who worked full-time as a software engineer while working on a Masters in Statistics part-time at Georgetown University. He then left his job to pursue the Biostatistics PhD at Harvard. His undergrad degree was also in Biology, not mathematics or statistics.

If you have strong performance in real analysis and can get great letters of recommendation, I think you could have a good shot at Biostatistics or Statistics programs. Here is a newly hired Assistant Professor of Biostatistics at UCLA who majored in German and Classics at UC Berkeley as an undergrad but then switched to Statistics/Biostatistics after his BA: https://andrewjholbrook.github.io/ I also personally know people who majored in biology, psychology, economics, and journalism as undergrads but switched to Biostatistics or Statistics later on. Often times, these folks had to take the math prerequisites (the Calculus sequence and linear algebra) and then got a Masters first before going on to earn their PhD. But in your case, I think you could probably just apply directly to PhD programs since you will have taken all the math prerequisites as a non-degree seeking student by the time that you apply.

If you might be interested in biostatistics, I think you could also get into really good biostat PhD programs. I could see you getting into somewhere like University of Pennsylvania Biostatistics and biostat programs ranked higher than that as well.

Are you domestic or international? If you are domestic, I think you could probably add a few more schools in the range of NCSU through University of Minnesota and one other "reach" school besides UW. With your profile, you can probably get admitted to Rutgers, ASU, and UCSB, so I would recommend applying to more schools around the range of TAMU. Just a note though: NYU has only 7 PhD students total, and they appear to all be international, so it might be extremely difficult to get into NYU.

If someone is going to transfer programs, I would recommend that this be done in the second, or sometimes, third year. There was one student in my PhD program who had spent three years in a pure math PhD program before enrolling in the Statistics PhD program at my alma mater. Transferring right away after one year seems odd and actually is awkward. Reapplying to a new PhD program in the second or third year isn't that awkward or necessarily unusual. Most professors are understanding of this, especially if the student who intends to leave hasn't even started research (then they have no personal stake in the matter). They might only have a personal stake if the student is one of their own advisees who is in like, their fifth year of study and wants to transfer (but then transferring instead of just finishing would be kind of strange).

You haven't told us why you are having serious second thoughts about the program you're joining in the fall. Either way, if you decide to transfer, you will need to explain why you are reapplying to a new PhD program in your application, regardless of whether you transfer in your first, second, or third year. And you need to make it clear that you're not doing it because of academic difficulties at your current program but because your needs would be better served at another department. Maybe one of the other faculty members on this board who has served on a graduate admissions committee can give their input about how a PhD application from someone who has only been in their PhD program for one semester is perceived. There probably aren't a super high number of such applications, but how do you think a GAC would perceive this? I'm sure it raises a lot of questions, but maybe it's more a matter of how you justify it.

You can probably find this information on the program websites, or barring that, you can ask the graduate coordinator what their policy is. Many programs like University of Michigan and NCSU allow entering first-year students to take the first-year qualifying exams (based on Casella-Berger and applied regression/design of analysis) upon arrival if they have already earned a Masters degree. And if they pass them, then they do not need to repeat the coursework. I would bet on taking at least one year of courses though, regardless of previous Masters or not. But you might be able to trim off a year of classes.

I assume their letters were from professors they took classes with. For PhD admissions, the mathematical ability to handle the courses and "research potential" are the most important things (analogously, if you apply for TT faculty jobs, the potential to meet their tenure standards is what departments will be most concerned with). If you were performing well in the classes in your ~top 10 program, then most programs will probably think you have both.

The Masters is just nice so you have "something to show for" your time at your current institution, but I suppose it isn't essential to have it if you've already been accepted to a new program. ? The two people I know who transferred from my PhD program had not started doing research under a PhD advisor when they reapplied/left -- they only had coursework. Some research experience might be helpful for your application, but I wouldn't say it's necessary if you've already been doing well in your classes at a reputable program.

I don't think it should be an issue to transfer if you've passed your qualifying exams. At my PhD program, I know there was one person who transferred to University of Wisconsin Statistics and another who transferred to UNC Biostatistics. However, they both completed their Masters before starting at their new institution. If you're going to do this, make sure you're getting a Masters out of your current institution. Professors are busy people, so unless they had a very strong investment in you staying, they probably aren't dwelling on it and it's probably not as "awkward" as you think. People Master out of PhD programs all the time. Also, I wouldn't recommend trying to transfer in your first year. That definitely looks suspicious. It is safer to do so in the second year if you're going to do it at all. If you've passed your qualifying exams and done reasonably well in your coursework, you can mention those things in your SOP (or have one of your letter writers mention that), so the new program knows that your decision to transfer is not due to academic difficulties but due to evolving research interests, etc.

See the above profile. In Biostatistics, I am not aware of any programs that 'strongly recommend' the Math Subject GRE, let alone require it. It's just not that useful for Statistics, plus so many international applicants (particularly from China) are able to score very well on it, so it rarely "makes or breaks" an application (except at Stanford).