Geococcyx

I'm not the person to give the full story, but they definitely have some quite strong professors -- look up Cun-Hui Zhang for an example (he does frequentist regression research, if memory serves). You could do well there, although with lower ranked programs who you work with is usually a little more important than at higher ranked programs.

My math background was pretty similar to yours; it was pretty light, and I'd taken linear algebra my first semester of undergrad as well. Bayessays pretty much hit the nail on the head for me, since I got waitlisted and admitted at NC State, and was likely waitlisted and admitted at Duke. You went to a better school, and did better in Real Analysis, so you might end up being marginally stronger. Even so, biostat programs are definitely places to look -- just as an example, UNC biostat has Dr. Kosorok doing machine learning, and lots of statistical genetics folks that might mesh well with you. Beyond that, I'm sure folks can help you with specific fits as needed. I recall Cornell being good at high-dimensional statistics, which meshes pretty well with your topics, so it makes sense that your professors and Bayessays recommended them. Wharton takes very, very few domestic students each year (and very few students in general), so they're a tough for most people to get in, hence why applying there as a reach might not be as helpful as you'd like. As to your other questions, you don't need to retake the regular GRE. Unless you really think you would knock the Math GRE out of the park, and it would be a really clear benefit to you, then I wouldn't bother either (provided you aren't applying to Stanford, of course, in which case you'd need it). The theoretical/applied split is hard to tell, but you can pick up some ideas of it in past posts on these forums, or else just by asking. UNC statistics seems to have a reputation as pretty theoretical, and being good at stochastic processes and whatnot. NC State would tend to be more applied. Places like Columbia and CMU probably fall somewhere in the middle, and Duke might fall a little more applied than those two. Ultimately, it's a case-by-case basis, though, and just because you might go to UNC statistics doesn't mean you couldn't work with somebody on, say, Kalman filtering and data assimilation for numerical weather prediction. (final note: Bayessays has been around rather longer than I -- if they or another older poster argue with anything I say, I'd take their answer above mine)

In general, you shouldn't consider no response a rejection so early. Maybe not immediately, but if some folks turn down interviews, then they may invite more to keep similar numbers of interviewees. Alternatively, if enough people turned them down after interviews then they may look at the top people they didn't invite out. Those are possible even if they hope to interview anyone they're interested in and they invite everyone at once, which as you note aren't necessarily the case.

I think(?) you'd want to go through the director of graduate studies rather than the admission committee, but you're welcome to ask them. I should note for anyone else that unless the school specifically asks for your fall grades, you don't need to send them, so it's only really worth sending fall grades if they're to your benefit.

It's probably a little lower than where you're aiming, but I think Duke stat has good people in TDA (Sayan Mukherjee) and privacy research (Jerry Reiter), although my lack of knowledge of differential privacy makes the second point a bit speculative.

Yeah, you'll get in somewhere -- as I recall, Virginia's just a small program, so they're harder to get into (just like NYU, Northwestern, Brown biostat, etc.) because of their small size. I got into Illinois last year without taking the Math GRE, with roughly the same math background from a worse university. There's no reason to take the Math GRE unless you're applying to Stanford. Maybe it helps at Columbia, UChicago, and UC Berkeley, but I know the first two accepted people last year who didn't take it, so it's not a big deal. Bayessays posted while I was writing, and they are correct about your chances. I think you should apply a little above Duke, and maybe a little above CMU, but I do usually suggest people apply higher than bayessays does -- that's your choice, and ultimately is a factor of how much money you have lying around to throw at applications. In any case, you're definitely aiming too low right now.

Which classes, if any, have you been writing proofs in? I would guess that you might have had that in honors linear algebra, and maybe(?) stochastic processes, but I'm a little surprised to not see something like an intro to proofs, advanced calculus, or abstract algebra listed -- not that there's anything wrong with that, of course, I'm just surprised that (say) TAMU's math curriculum wouldn't have you taking one of those by end of your junior year. Then again, I'm not very familiar with your school's curriculum design, so maybe I'm just misunderstanding. I don't know that your research would really be an issue. The ranking of your school in the overall USNews rankings matters, of course, but public vs. private shouldn't really matter beyond that. As a side note, you mean top-50 out of publics, as opposed to a public school in the USNews top-50, correct? Those might help me decide, but bayessays is probably right about your chances, so you should follow their advice.

You should get letter-writers that convince the schools that you are good at high-level math and might be good at research. As important as TA's may be at many universities, you're not going to get into a PhD program because you'd be a good teacher. Even so, with my apologies, I don't think you would get into a statistics PhD program if you applied with your current profile. Your quantitative GRE score is too low, as you know. I'm guessing from your master's program title that it didn't include real analysis, and unless you took that and did very well (in grad school or undergrad), you're gonna have a really tough time getting into a PhD program (and you'd need a 163 or so on the qGRE to go along with it). If you want a PhD in statistics, you'd want to take real analysis and a probability/statistical inference/mathematical statistics sequence and do really well. That might be in a statistics master's, or maybe as postbacc classes, but that would be a start, along with a higher qGRE score and maybe finding some research experience along the way. I'm hardly an expert, so if an older poster contradicts me then you should listen to them, but this is probably a good start. I'm sorry in the present, and I wish you good luck for the future.

Well, let's start with this: if you could get your quantitative GRE score up to 164, I think you'd have a decent chance for most schools except the very top (Stanford, maybe Cal, Harvard, and UChicago?). I might be overstating your chances at each place, but I think you'd have a decent chance to get accepted at UWashington or a school of that level. I can't really judge your research, but you might be in even better shape if it's particularly good work. As a note, you don't really have to submit a Math GRE score except for Stanford, so unless you're particularly interested in going there, you probably don't need to worry about that. With your current GRE score, it is going to hurt a little bit, since sometimes having those high math grades but low quantitative GRE scores makes people wonder about the rigor of the classes/grading scale thereof. At a top-50 school, though, that won't be a big concern -- even less so if your school has a strong math department. As such, it might hurt you a little bit by comparison to other applications, but I feel like some of those schools would still be inclined to "take a chance" (if we can call it that) on someone with your grades. I think that UWashington to Penn State area, maybe down to UCLA or so in the statistics rankings, is a good place to concentrate applications. Overall, I can't really recommend schools for your research interests (although UNC statistics seems a likely candidate), but I think you can apply to more of the elite programs you're interested in. I'm not the expert on profile evaluations or this kind of math, though, so listen preferentially to the more senior posters if they drop by.

Just to clarify my lack of comment: I largely agree with omicrontrabb for statistics PhD programs. I am, however, a little less sure whether all students at unranked/low-ranked biostats programs come in with real analysis (you mentioned an interest in biostatistics research, so I've imputed an interest in biostatistics programs for you in addition to statistics). I also suspect that having a good coding background might in some cases predispose programs to take chances on people with lesser theoretical backgrounds. Even so, you're gonna have a pretty hard time getting into a PhD program without real analysis. At my school, discrete math was very similar to our intro to proofs class, and they could count for each other as prerequisites for math classes; if yours was similar, you might be in a bit better spot that omicrontrabb suggests, but not by a whole lot. Beyond those comments, omicrontrabb's recommendations are worthy of thinking through. Also, I'd imagine Dr. Minin would be happy to advise a former student regarding grad school.

Is there any way you can take real analysis, or otherwise some intro to proofs/sequences and series class this upcoming quarter or two? That would probably do you a lot of good for your viability at many schools. For now I'll defer on giving you a judgment on what schools you'd have a chance at, since I'm not confident in my accuracy, but taking real analysis or a similar class should probably be a top priority if an option for you.

Most people don't have undergrad stat research (particularly methods research, which is the most important kind for applications). The math REU is a definite positive for you. Your research background shouldn't be a problem. Re: letters of recommendation, if the two professors you have close relationships with taught proofs-based math (esp. real analysis/measure theory) or statistical theory, or else were your advisor at your math REU, then you might be better-served by choosing a professor you do know well from a class that isn't strictly math or statistics. I chose a professor from a genetics class because I thought it would tie-in well to how I was selling myself to departments. If you want to talk about being a good statistics communicator, maybe get a professor from a communications/film/theater class that you know pretty well. Maybe you like philosophy of science, and have a relationship with a professor from intro philosophy. Maybe you like genetics or biomarkers or engineering statistics or forensic statistics -- choose accordingly. If those two professors aren't speaking to your ability in math/proofs, though, then my best advice would probably be to talk to them in-person, maybe talk about a class project/time you went to their office, and have your CV, transcript, and at least draft of personal statement/basic angle you're selling yourself through ready to talk to them about. If you got a particularly high grade, that might also be something to note.

I don't know about "most", and these days some schools will partially discourage it by requiring a master's thesis, but it seems fairly common. I'm quite confident that NC State and Duke let people master out, and I know Stanford at least used to. Plenty of others do, I'm sure, but as you might imagine I didn't research it at most places.

Most of what I've heard on hear (I believe at least in part from bayessays, who might weigh in themselves) is that a master's won't really help that much barring special circumstances, e.g. changing fields and taking real analysis for the first time. That is not an issue with you. If you're wary about a PhD, then you are in luck -- UChicago has one of the two best statistics master's degrees in the country (along with Stanford), and unless all of your low grades were in real analysis or something, I don't have any reason why you wouldn't get in. Really, I think you can apply to most of the top PhD programs and work from there. If you don't like or aren't capable of research, then you can choose to master/ABD out, without having to pay back loans from your master's tuition. You'd have to do pretty well on the math GRE for Stanford to accept you, I think, but given that I didn't have a chance at places that highly-ranked, I don't think I can comment very accurately on your chances beyond just that you should do considerably better than I.

You're applying to a slate of schools that seems pretty similar to what I applied to, and I think your profile is stronger than mine, which would suggest that you should probably be able to apply to these schools with a good expectation of some choice (although, as a warning, I do feel like a lot of applicants on here are stronger than I was). Maybe you could aim for a slightly more top-heavy approach to provide more high-tier choices, or at least consider applying to a couple high-ish places in addition to what you already listed above. Bayessays knows more than I do, so I don't intend to disagree particularly, but I think you'd have a decent chance at Duke at least -- since they're pretty exclusively Bayesian, they use a lot of MCMC, and I think that might predispose them to liking people with physics/physics simulation backgrounds. To echo bayessays, the higher ranked biostat departments will have the theoretical background you're looking for. I've usual heard that two of the top biostat programs require measure theory -- Washington (Seattle) and UNC -- so those might be ones to consider as well. That said, they would probably all let you take those more theoretical courses if you wanted, so that shouldn't be a reason to remove Johns Hopkins or anything.

If it eases your feelings a bit, I had a professor who mastered out of a top statistics school and then got admitted for a PhD at another top statistics school immediately thereafter. You would have even less to worry about than they did since you changed fields, and clearly it wasn't a problem.

I'm not very knowledgeable about master's admissions, but I don't think there's really any reason to worry here. You have a 4.0 as a math major, and you took pretty much every class (besides measure theory) that people would be looking for in a PhD applicant(!), plus the GRE scores to go along with it. I'm aware that school "prestige" means rather more than I feel it should in applications (at least for PhD's), but even so, you have the profile of a student who was so good at a less prestigious school that you'd be considered anyways. In any case, master's programs mostly care about ability to do the classes, and you've clearly demonstrated an ability to complete difficult math-based classes well. EDIT: This is to say, I think you can apply pretty much anywhere you want.

I'm not great at evaluating international students' profiles, so don't put too much weight on my thoughts -- just a word of caution. I'd be inclined to take your professor's thoughts as they are; Ohio State and Florida seem like reasonable places for you. The B+ in Calc III shouldn't mean much given the time and classes you've had since then, and the B+ in probability shouldn't mean too much either; those aren't even bad grades, really. I do get the sense that you aren't really as concerned as you say, though, since you largely are considering schools higher-ranked than Ohio State and Florida. Looking at some schools above that level is a good idea, of course, but I'm not quite sure how good your chances of getting into Columbia Stat or Brown Applied Math would be; again, I'm not good at evaluating international students' profiles, so that is a bit of a shot in the dark for me. I'd tell you to stay more in that range of Penn State through Ohio State and Florida, with maybe a couple above it (like NC State and Duke, as you have), plus a bit more coverage below Ohio State and Florida for additional choice later-on. Maybe your research experience would bump you beyond what I'm saying, but I just don't really know how to judge that very well. I'm not an ecologist or anything resembling it, so my area knowledge of it is completely lacking. We did recently have a thread of schools that work on environmental statistics, which might be helpful -- schools that came up there were NC State, Ohio State, and Oregon State. Additionally, Cyberwulf at one time mentioned some biostatisticians who do environmental statistics, such as Roger Peng (Johns Hopkins Biostat), Francesca Dominici (Harvard Biostat), and Amy Herring (Duke Statistics). I think the University of Georgia has a good ecology school (one of the Odums worked there), but I don't know that their statistics department is very involved in that work; even so, you could look there. I think UNC is tied to the Odums as well, so you might look there as well (along with Florida, so Wikipedia informs me). Duke Statistics also used to have a professor who explicitly worked on statistics in ecology (Alan Gelfand), but it seems like he retired recently. Hopefully this helps, and if you'd like, it wouldn't be a bad idea to consult with your professor some more too to make sure of your choices (their time and willingness allowing).

I second bayessays, I did fine in admissions even though I was taking real analysis at the time (and hence, most schools didn't have any idea whether I'd done well in it or not). If you want a PhD, then you should be able to start on it next year!

I'm very much not the expert in how admissions for international students work, but I'll respond in the interest of giving you a timely(-ish) partial answer. Responses to your remarks/questions. 1. You didn't give us the grades for your classes, so how much that one lower-GPA semester would vary greatly depending on which math classes you took during it and what your grades were in them. That said, the one I was worrying about most was Real Analysis, and you have an A+ in that, so I doubt it'll be an issue unless you got a C+ in linear algebra that semester or something. Even then, your profile's pretty strong, one semester averaging a B+ probably shouldn't be something to really worry about. 2. I'll get to the school list below. Most stats classes aren't going to be a big help -- a grad level statistical inference class would probably help, if not that then an undergrad statistical inference class or the grad level stochastic processes class you mentioned. 3. Again, below. 4. I don't really see an issue with asking your Real Analysis professor. Presumably your other recommenders are your REU advisors or whatnot, that cohort of professors should be good. 5. I can't comment on those schools' research areas. OK, now your school list. I'm not much of an expert in Math PhD applications either, so there's a limited amount I can comment on that. That being said, your statistics list is really strong, and while you're a good enough applicant that you might get into some of those schools, I'm guessing you would want some safer choices in there too. Part of that is that my limited impression of Math PhD programs is that they're really hard to get into, and part of it is that if all of your safety schools are math programs then that doesn't seem optimal for you -- if you're not sure you want to do math and are looking at statistics as an option, it doesn't make a lot of sense to strictly aim for some of the most difficult statistics programs to get into. It's obviously fine if you would still probably prefer to do a Math PhD instead, but insofar as you want to explore statistics, it'll probably benefit you to apply to a few schools in the NC State/Texas A&M/UCLA range just for the ability to choose once you've visited them and make sure you have a non-math option.

Well, this is probably a situation in which the Summer Institute for Biostatistics/Summer Institute for Training in Biostatistics (the names are inconsistent), otherwise called SIBS, might be helpful going forward. That's a program funded by the NHLBI that teaches people the basics of biostatistics, and gives you connections to faculty at a generally pretty good biostatistics program. That said, if you are applying to PhD programs this upcoming school year, you clearly can't apply to SIBS since it started about 3 weeks ago, and ends about 2 weeks from now. A good place to start might just be to go to a biostatistics PhD program website -- sometimes they provide some basic explanation of the subject, but they also list faculty from that school that work on that topic, which lets you go look at those faculty members' publications and read more about it yourself. Some examples of these lists of research topics are at Emory (https://sph.emory.edu/departments/bios/research/index.html) and UNC (https://sph.unc.edu/bios/bios-research/).

As a heads-up, I have maybe a fifth of the experience evaluating people's profiles as some of the other most likely posters on here, so don't take my word as gospel. Even so, I hope I'm not too far off. Overall, I'd say that you should be applying in roughly the range I did (as you can see in my signature) -- potentially better. I had a better GPA from a similarly good state school, and really high GRE scores that might have oversold my abilities. Even so, though, you clearly have a better math background -- when I was applying, I had taken 2 proofs-based math classes, and was currently taking real analysis. As it is, I would think you could take shots at most of the top 10 biostatistics schools (via USNews grad school rankings); the top 3 (Harvard, Johns Hopkins, Washington) are tough to crack, so those would be reaches, but places like UNC, Minnesota, UCLA, Wisconsin would probably be good options. Some of those might still be reaches, but I feel like if you apply to several of them, you'll stand a good chance of getting in at at least one of those 4-10 range schools. If you're still in school this upcoming fall, and are taking more real analysis/measure theory or a grad-level statistical inference class, then you'd be more likely to get into those top programs. If you're interested in both probability theory and mathematical biology, then you might also consider schools with combined statistics/biostatistics departments, like NC State and Illinois; these might fit your interest profile a little better than just a biostatistics department. You'd still probably want to choose a couple safety schools a little below where I'm talking about in the rankings -- my only particular note about this is that a lot of the biostatistics programs below the ones I listed are pretty small programs (Yale, Brown, Duke, Vanderbilt) that may not make for good safety schools just due to how few people they accept each year. That said, I've leaned pretty heavily towards biostatistics departments in this analysis, partially because your math background will probably be a little stronger in that application cohort. That doesn't mean you need to go to a biostatistics or combined statistics/biostatistics department to do biological applications. For instance, the Duke statistics department has several top faculty members who work on biological applications, Columbia statistics has several people who do neuroscience research (as I recall), and you can work with Michael Newton (an award-winning statistical geneticist) as either a Wisconsin statistics or biostatistics student (as far as I know, anyways). I hope this helps. If I've erred in evaluating you, I hope I've underestimated you -- clearly I did fine in applications with a limited theoretical math background, and I would hope non-major GPA and GRE scores wouldn't mean much.

Can you give us an idea of roughly where your big state school is on the USNews undergrad rankings (for example, Clemson and Texas A&M are tied at 66th)? Admissions committees will take your record a bit differently depending on whether you go to, say, UCLA or Mississippi State. Regardless of your school's ranking, though, I don't think there's any reason you'd need to do a master's degree first. We can work on target grad schools once we know more about your undergrad school, but do you have any areas of research you're especially interested in? Are you trying to stay in mathematical biology, or do you have interests in social science or probability theory or time series too? Any sort of preferences geographically, or in the size or characteristics of the programs?

Maybe it's not a big deal, since Stat PhD didn't mention it, but could you potentially take Real Analysis in the fall instead of the spring? I'm not sure it would do much, but that way if programs were on the fence about you, they could ask you to send your fall grades and look for a high Real Analysis grade to convince them. That also works for programs with application deadlines in January or later (Texas A&M and UConn are two places that come to mind), plus Wisconsin asked everyone for their fall grades this last cycle from what we could tell.

My background isn't as close to yours as bayessay's is, but maybe I can be of some assistance via my own stream of consciousness (which turned out to be super long, sorry about that). Keep in mind that I also have substantially less experience in this than bayessays and some other folks on the forums. My personal opinion is that you'll probably have an easier time strengthening your recommendations by choosing different recommenders rather than by getting a master's degree. Keep your strong letter from a research advisor, of course, but I feel like you have options to work with for the other two spots (or maybe 1 -- I don't know whether it's frowned upon to only have recommendations from professors once you've been out of school for a few years, although I personally doubt it would be a big deal). I don't think people would worry too much about your math ability given your A's in real analysis, but it's not like it would hurt you to have a letter from one of your real analysis professors saying that you did a really good job in their class, or perhaps from your measure theory professor detailing how you improved from C+ work to A/A- work over the course of the semester. At the same time, having a breadth of math background gives you the latitude to choose a professor from a less mathematical background than econ for a letter if you want. If you ended up taking a philosophy or literature class that you really enjoyed towards the end of your collegiate career, that professor might be a reasonable choice if you talked to them a lot and demonstrated creativity, analytical thinking, and good written or oral communication skills. Your computational linguistics class might provide an opportunity like that too; I'm woefully short on knowledge about that specific subject, but classes like generative syntax and similar linguistics courses make for great demonstrations of critical thinking, even if they don't seem super relevant. One of my recommendations came from a genetics professor whose class I took despite it counting towards no degree or graduation requirement precisely because we spent a lot of time thinking about comparisons of models, how to infer causality, and that sort of gedanken work; I'm sure the genetics background helped for biostatistics applications, but that surely wasn't the only useful information committees got from that letter. I'm personally a fan of taking a few completely out-of-degree classes on a lark during your college career, and I think those classes can be great opportunities for letters of recommendation. I find that taking classes in new subjects makes me more excited for that class, encourages me to develop more of a relationship with the professor through asking lots of questions, and will probably be in an area that many other applicants don't have experience in, which will help you stand out during the application process. If you did that in college, great! If not, the same thought process might apply to work -- maybe still have the Econ PhD write it, but you might point him to some experiences of good management, planning, and critical thinking on your part. If nothing else, that might convince a PhD program that you'll be more disciplined than some younger applicants, and in a PhD program that's not a bad impression to make at all. Ultimately, I think you're too hard on yourself -- my recommendations were a research mentor like you, plus two professors that I'd had all of 1 class with each. Those were small and talkative classes, of course, but coming from a large state school my baseline for a small and talkative class might still be rather larger and quieter than yours. Choose folks who can say fairly unique and good things about how ready you'll be for graduate school regardless of the subject area, and then balance that with a need to have some research and math background testimony in there. You almost assuredly don't need to go to a master's program just to achieve that.