StatsG0d

I think any program outside the top-3 (UW, JHU, Harvard) is plausible for you. Based on your history, it looks like you're really into theory, so I would say the only biostat programs that would be a good "fit" would be those top-3 and UNC. The other biostat programs are less focused on theoretical training. I'll close by mentioning it does not really matter much where you study. If your advisor is well known and well connected, that will hold a lot more weight than coming from a top program with an unknown advisor. Publishing in JASA is really at least 50% luck as there is no way of knowing how groundbreaking the project will be when you start it. For that reason, your advisor's recommendation will be the biggest deciding factor in where you get placed post graduation. Take a look at @Stat Assistant Professor's recent post, where he talks about someone that came from an unranked program and is currently a professor at UF publishing in some of the best journals.

Sorry, I misunderstood "top-100 Asian" as "an Asian university in the top-100" If your school isn't necessarily renowned for producing top quality graduate students (e.g., Peking, Tsinghua, ISI), the best strategy is to find out where your school has successfully sent PhD students, because they are already familiar with how competitive it is and know how the graduates of these universities fared at their programs. Perhaps you can talk to some professors / the chair of the department. I am sure that they know where students have gone. These schools should comprise your "target" schools. For reaches, I think you have a small but nonzero shot at getting into a program ranked 15-20 in the USNWR for statistics programs (ignoring biostatistics). Here is a list of the rankings separated by stats / biostats (make sure you scroll down to Marmle's post). For "safeties", I think any school ranked 40 and below would be some good choices. At this level, there's not much difference between a school ranked 40 and a school ranked 60, so I would say select some schools that have faculty in your research interests and that's in a desirable location.

I think your profile is strong. Do you know if your institution has successfully sent many applicants to US PhD programs? Based on the rankings, I'm assuming you have gone to one of Yonsei, NKKU, or USTC. I'm not familiar with the latter two and only slightly familiar with the first.

That definitely changes things. You should post your entire profile using the template that everyone uses so we can give a proper recommendation.

Just because your average GPA is lower, does not mean you are less qualified than the other applicants for two reasons. First, assuming that the distribution of GPAs is symmetric about the average, 50% of admitted students have GPAs below the average (in reality, it might be a little less than 50%, but the point is many applicants have to have below the mean GPA). Secondly, you have to realize that these average GPAs are the average of applicants from all schools across all majors. An applicant who received a 4.0 from the University of Alabama in engineeering will be viewed very differently than an applicant who went to UC Berkeley with a 3.66 and who took very advanced math courses. The reality is that the vast majority of MS stats students will only have taken the bare minimum math courses for admission, or maybe slightly above. If you wanted to, you could probably apply to some PhD programs, and I can guarantee that you would get accepted to some.

Sorry to say, but I do not really see you getting into any of those schools. You could maybe get into some biostats programs outside the top ones (e.g., Pitt) that are more focused on applied work, but few stats departments will take anyone without a real analysis grade, and you won't even be able to send a grade before the admissions cycle is up. If you're really set on a pure statistics program, you're likely going to have to take a gap year and get that GRE score up to a 165+ and at a minimum get that real analysis I grade. You have to view this from the side of the adcoms. Why should they take their chance on an applicant who has a shallower math background and GRE score relative to their peers? You do not want to leave any shadow of a doubt that you can do PhD in statistics.

I think it's naive to think that all (or even the vast majority of) biostats faculty members are fully engaged / committed to public health and/or biomedical problems. There are many professors who are in biostatistics simply because applications of their theoretical research area naturally fits in the realm of biostatistics (e.g., imaging, dimension reduction, historical data borrowing). This is probably mostly true for faculty members who were trained in stats departments rather than biostats departments. That said, I wholeheartedly agree with you regarding students.

I agree and I was going to comment something similar. I would say that you at least should have some interest in public health. If you're just in it for the math, you'd probably be happier at a statistics department.

They don't really teach any biology in biostatistics programs. It's only necessary to learn/re-learn bio if you're interested in statistical genetics.

I think we need a bit more context like is this for a PhD application or a master's application? How is the rest of your profile? etc.

I think @bayessays hit the nail on the head. I would hesitate to call statistical genetics applied. The vast majority of work being done is methodological--coming up with new methods to find genes, etc.

Well, the SPH department only has 1 full professor, 2 associates, and 3 assistants. Of those, only 4 were trained in stats / biostats departments. Their Med School department is much larger, having more senior faculty. But if you look at their publications, they are more focused on public health / epi than methodological work. They do publish in high impact medical journals like NEJM. Among the ranked biostatistics programs, I would say they aren't really comparable. It depends on what you want out of it. If you want to work in industry, it likely doesn't matter too much. If you want to work in academia, I think it would be difficult.

I agree with @Stat Assistant Professor. Students have been pushing to modernize curricula, but it's difficult because professors are always concerned about prestige or rigor. Some topics I think should always be covered are: Bayesian statistics (becoming more and more used in practice, even being picked up by CS people) Computation / simulation (preferably in C++ / Python and on Unix servers) Machine learning / nonparametric statistics (may be a buzz word, but it gets you jobs) Missing data (very common in practice) Some topics I think can be tossed out, that are typically required: Measure theory (useful for many people, but not for all) Decision theory (hardly ever used in practice) Anything concerned with unbiased estimation (UMVUE, etc.--most practical estimators are biased so who cares) I do think UMP and UMPU tests are important, albeit boring, at least for biostatistics. Drug approval ultimately depends on having a significant p-value, so you def. want to have power.

The classic book for linear models in a PhD program is Christensen's Plane Answers to Complex Questions.

Agree. 168 won't raise any doubts at all.

I disagree with that. The training in the vast majority of stats programs is very theoretical, especially at the schools you've mentioned. You can do something applied for your dissertation, but it will likely need at least a bit of new methods / theory.

Others may disagree, but I think you could go almost anywhere. What's going to be important is that your advisor is well connected and can hook you up with a good postdoc afterwards. The postdoc will be the real difference maker when you're applying for faculty positions. For industry, it really doesn't matter what you do / with whom you work. There might be some positions that require some expert knowledge in some subfield, but the vast majority of jobs in pharma / FDA really only require advanced knowledge of standard statistical methods and a solid understanding of clinical trials and issues that arise with regulatory agencies (e.g., multiplicity). You'll definitely want to know Bayesian statistics, as it's becoming more and more used (for trial design / sample size planning, if not for the analysis of the data). I also recommend you become an excellent programmer, as that will set you apart from other candidates. Any PhD graduate understands GLM, but few could program their own algorithms that are very efficient and accurate.

If you're interested in biostat, you should revise your school list. UNC STOR does basically zero biostats. You could apply to nearby NCSU, which has several people in biostatistics. In general, the larger state schools that do not have standalone biostatistics departments (e.g., NCSU, TAMU) tend to have some faculty working in biostatistics. Now, onto your profile evaluation: I agree with @Casorati--your success will depend somewhat on your undergrad institution. However, you do have a master's degree from a North American master's program with a great GPA and from a prestigious program. This will help you out a lot compared to other international students. You also have a perfect GREQ which is great. To me, I think it will mostly depend on your letters. I think you should spend some time with your letter writers who might not be as strong and just speak to them about your interests in your passion. You want them to be able to write that you'll be a very successful students and you can do math very well etc. For international students, biostatistics is more difficult to get into than statistics departments because they are more reliant on NIH grant money, which can only be used to train US Citizens / Permanent Residents.

I feel like UCLA is deceptively somewhat difficult to get into for both stats and biostats. This is anecdotal, but I applied to both and got waitlisted (ultimately decided to withdraw) from both despite getting into several schools ranked much higher. Maybe they'll prefer in-state for tuition waiver purposes. That said, I feel like your math background is pretty shallow. I'm not sure how an online analysis class will be viewed to the adcoms. I think you might have a shot at the other schools that are not UCLA, but I think UCLA will be pretty tough. At any rate, you can apply to both stats and biostats at UCLA / other schools to broaden your chances.

There's really not much else you can do. You have a very strong profile. Apply to a couple top-10 schools and target schools ranked 11-30. I am certain you'll have plenty of choices.

1. There are many students who do not have a big stats / probability background that are accepted to PhD programs. Usually the problem would be the opposite--people apply majoring in statistics or another quantitative discipline but do not possess the required math background. Your background is fine, and I think the motivation you mentioned already sounds good for the SOP. 2. Few applicants have meaningful research experience. I think simply mentioning that they are on arxiv will be more than enough. 3. You definitely have a good shot, and it's better to come form a lower tier US university than an unknown international one (even though the latter might be more rigorous). It's a lot tougher as an international student, but I think your GPA and math background are enough to get accepted to some top-30 programs. There are a lot of schools that tend to recruit many international students (UF, FSU come to mind). I'd be surprised if you didn't get into at least one of those. 4. I don't think it sends a bad message. What matters is that your letter writers can attest to your ability to do math. It's great if it comes from a stats professor, but I would rather have 3 letter from people whom you know are going to write something positive than 2 positive and 1 neutral.

Any successful PhD student has likely struggled at several points in their training. My first year (Casella-Berger year), I barely managed a 3.5 while my peers were getting Mostly A's and A-'s (for the record, it's pretty hard to score below a B in grad school). After the first year, I was getting better grades than many of my peers who crushed me in the first year. The point of the PhD training is to tax you mentally so that you can start to mature mathematically. I personally do not think grades or how well you do on the qualifying exam will make you a good researcher. It may be different in some old school professors' eyes, but I think most people these days view the qual / courses as a means to an end. At some point in your career as a graduate student, things will start to click together. And it's very possible you'll never see/use measure theory stuff ever again after taking it. One of my peers was a bio major in undergrad, and ended up receiving the highest score on the theory portion of our doctoral exam. They had little/no previous exposure to real analysis. They are an extremely hard worker, so there's that. But all this to say, I think it's extremely possible for a bio / CS major to be successful in a statistics / biostatistics PhD program, albeit maybe the latter more than the former.

Check out @cyberwulf's post stickied to this forum. In it, he indicates that research interests matter very minimally.

You do not have to commit to doing the PhD. You can enter the PhD program and leave with a master's. It's becoming more and more common. Also, 4 years for a PhD (especially without a master's) is very fast. It's more like 5-6 years.

Your profile is strong, but based on your research interests I think you'd be better off being placed in economics departments or some of those quantitative methods in social sciences (QMSS) PhD programs (e.g., at Columbia or Michigan). There are some people in biostats and stats doing causal inference research, which is related to economics, but overall I would say econometrics folks have the edge. Econometrics and statistics, while related, can really be quite different.