Geococcyx

Have you taken any math classes (e.g. Calc 1, 2, 3, Linear Algebra, Real Analysis, Intro to Proofs, etc)? If so, what were your grades? Also, was your GPA consistent across the years, or did it rise or fall a lot, and if it did change a lot, were there any reasons for it? That's a tough GPA to work with, but it would help to know this info. I'm just another student, but I suspect you'll get better advice from other people if you answer some of what I asked.

It looks like you've been around the forum for a bit, so I'm not sure how much other information I can give (especially given that I'm just another applicant), but if you'd like some allaying of your concerns, maybe I can still be useful: 1. Most people don't really know their research interests that well at this point in their careers. I had a few application areas, and maybe one vague theoretical area, that I expressed interest in via my personal statement, but you could probably get by with less (based on my experience at visit days), and as long as you aren't amazingly narrow, having more developed interests shouldn't really be an issue either. I'll echo some responses to your previous posts and say the best ways to learn (aside from asking about specific schools on the forums, which sometimes works) are to look up some top professors at schools you're interested in and look at their papers. You don't have to read the whole thing, but at least go through some abstracts. Listed faculty research interests work too, although the quality of such postings varies pretty widely between departments in my experience. 2. It wouldn't hurt to work through Casella and Berger, but if you have strong grades in real analysis, measure theory, and other proofs-based math classes, then I doubt it would make you look that much better. I'd assume going through Casella and Berger would be more helpful in preparing for the grad program itself, rather than the application process. 3. I don't recall such online courses really meaning all that much to admissions committees due to vagaries in their grading, so I doubt it would be an issue. Frankly, plenty of people (myself included) are mostly at a loss with regards to computer science topics like algorithm design, software development, and whatnot, and generally it's not any sort of drag on our applications. It's a little hard to know how well you'll do without GRE scores (or at least SAT/ACT proxy measures of standardized test ability) and grades in your analysis/measure theory/probability sorts of classes, but if I had to guess you will probably have a quite strong profile -- maybe not tip-top level, but certainly stronger than mine, particularly if your math publications are in analysis or some area relevant to statistics. I wouldn't stress it too much.

I'm just another student, so be cautious taking my advice, but I think you're probably fine --more often the issue with stat applicants is that they haven't taken Real Analysis, so your pure math background is more of a positive if anything. Maybe biostat programs would care a little more, but probably not a lot. Most applicants haven't done theoretical research, either, so your research might be a positive, although its hard to say from so little information.

I'm an applicant this year, so I'm just spitballing here. Your calc and linear algebra grades are perfectly fine, and those will be the most important for a master's. The Stat & Probability grade isn't great, and that might factor in, but I doubt your algorithms grade (for example) will really be an issue. Your GRE scores are quite fine, those should be good enough for anywhere. If you wanted a stronger profile for top master's programs (or maybe for a PhD program or 2 that you might apply to), then taking a real analysis or proofs-based sequences & series sort of class would help, but I don't know to what extent you'd really have time or money for that. My initial impression is that you'd probably have reasonable chances at the master's programs you mentioned here, but it might be difficult to get into Baylor's PhD program if you do apply. As I said, I'm not an expert, so if Bayessays or Stat PhD Now Postdoc or another more experienced poster comments, take their word over mine. If I may ask, what is your goal post-masters? As you're probably aware, some masters are geared more towards sending people into industry, whereas others can prepare you for industry but also cover enough theoretical background to make you a good PhD applicant, so that could inform any suggestions if future posters think you're under- or over-applying and want to suggest some appropriate places you might want to send applications.

@Tucker154 Stat or Biostat? For stat, they sent out a google doc a few days ago to see if anyone was ready to commit or decline, or else if they were leaning towards or against coming. If people are waitlisted for stat, I'd imagine they'd start to find out acceptances pretty soon, once more people respond to that.

If you are interested in mathematical statistics and stochastic processes, I would hazard a guess that UNC Statistics and Operations Research might also be somewhere good to apply to -- I imagine it would be a reach like NC State and Penn State, but it seems like you're applying to enough places that you could fit it in if you wanted to. Those aren't my areas, though, so you should check out the professors there before you make any decisions about it.

I'm just another prospective student trying to decide where to go, but I can probably provide a few details of varying quality (the actually good details are secondhand from older posters, but if I mention it now they may expound on it later): Rutgers has Cun-Hui Zhang, who appears to do work in model selection and high-dimensional data (or at least used to -- I don't think they have a Google Scholar profile, so I'm just seeing their biggest hits). Dr. Zhang has several really strong papers in Annals of Statistics, has something in PNAS, and so on, so I doubt you'd have too many issues in the academic job market if you worked with them. For what it's worth, my statistics adviser spoke pretty strongly in favor of my applying to Rutgers, and he was mostly encouraging me to apply to places like Duke, Columbia, and UCLA statistics, so Rutgers is outpunching its supposed weight, ranking-wise. (Credit to Stat PhD Now Postdoc for mentioning Dr. Zhang in previous comments about Rutgers) I've heard Rice is strong in financial statistics, which would probably lead to pretty strong industry placements. I don't know to what extent it would really effect the program, but I know Rice stat and MD Anderson biostat (which is right across the street from Rice) allow for some class crossover, which might(?) extend research areas if they allow collaborations with professors from the other school. Hadley Wickham is also an occasional lecturer there, if you care. As a mostly unnecessary note, I'm not sure that I'd rate living in Houston as more enjoyable than quasi-Sacramento or New Brunswick due to my impression of Houston's urban sprawl, but I can't claim too much expertise or experience about what it's like to live in any of these cities.

I recognize that I'm spoiled to already have this many replies to my thread, but since a lot of the debate has concerned how people choose their PhD advisers at UNC biostat, I thought I'd just post here to reopen the floor in case anyone had comments about the other aspects of either of these programs. If not, then thank you for all the information, everyone!

@Stats951 I'm interpreting "Asian" as meaning that they are a citizen of a country in Asia, rather than a citizen of the US (I'm not a lawyer, I'm unclear how permanent residents or DACA fit into this). Domestic students have a much easier time in the admissions process than international students, so domestic/international status is definitely important. I might be incorrect in that making that interpretation, though. Ethnicity might factor into grad school admissions in the form of a diversity statement, wherein some schools will ask how you contribute to diversifying the school/field -- I recall UCLA had this, maybe also Wisconsin.

I'm an applicant like you, so be warned about my advice. I don't really think I've ever heard someone on this subforum speak positively of contacting professors about admissions. As such, I'd assume that you probably shouldn't contact the interviewing professor or director of graduate admissions -- your getting in probably has a lot to do with other people declining or accepting offers and very little to do with what you've been working on since your interview. I would also think that your interest in the program would be assumed (people can withdraw from the waitlist if they want; you clearly haven't), so I do kinda doubt that you'd have to further state your interest, although I'm less sure of my opinion on that. I would also assume that a status update on April 1st isn't really something to worry about, but I'm also guessing that it depends on the school. If this is, say, NC State, and they have a gigantic waitlist that they outright state is unranked and they won't talk about it, then clearly don't ask. If this is a school with a really small waitlist and who wrote to you to confirm you were still interested in being on the waitlist, asking about your status might be a bit more reasonable, albeit I'm still not sure it would be an outstanding idea unless they offered to be open to such requests.

I'm just another applicant, so take my suggestions lightly. That said, I'm given to understand that epidemiology programs are more forgiving of light math backgrounds. I'm a little unclear as to what your endgoal is for your career (other than "not a SAS programmer in pharmaceuticals", of course), but depending on what you want, an epidemiology degree might be a way to get there. If you do want to stick with biostatistics/statistics, though, Stat PhD Now Postdoc's suggestions are probably the best. I'm pretty sure I outperformed my evaluations from this forum during this year's application cycle, so I'm not inclined to stop you from applying to a few lower-ranked biostatistics programs that you're interested in if you really want to, but I wouldn't spend too much money on that vs. enrolling in some theoretical math classes.

Looking at the syllabi for the 151ABC and 209ABC sequence syllabi, I think you'll probably end up stronger than Stat PhD Now Postdoc thought you might -- 151 appears to go up through Lebesgue integration. I can't really assess measure theory curricula, but 209ABC appears to cover the dominated convergence theorem, theorems by Fubini, Tonelli, Egoroff, and Lusin, Fatou's lemma, and Borel measures among many other topics, which are all probably beyond the more basic real analysis topics Stat PhD Now Postdoc thought the sequence might consist of. If you got through the 151 and 209 sequences with, say, all A's, I suspect schools would look pretty favorably on your analysis background. That part excepted, I'd take all of the previous post's advice and assessment to heart.

@little white in Stat I wouldn't feel quite so strongly about that, myself -- I doubt the professors would be particularly worse at choosing advisees than students are at choosing advisers. I just want to help ensure I end with a strong adviser and have a decent chance to get a paper or two in Annals, Biometrika, JASA, or such. Even so, both students and professors would have some say in an adviser/advisee relationship in general. It's possible advisers might tend to choose advisees more so than the other way around, but I kinda doubt you'd lose much autonomy.

I gather that Duke typically allows students to choose their adviser, so perhaps that is another benefit of Duke? Or does UNC have enough strong professors that their allotment procedure isn't an issue? I should've noted above, but just to be clear, I have no preference for or against Bayesian statistics.

I've already talked to several people on the forums about this, but I thought it couldn't hurt to open the topic up to other folks. It also seems like I'm not the only person who's been comparing these two schools, so I hope it might be a useful thread in the future. My research interests going in are primarily in applications to neuroscience, genetics, and climatology/environmental issues. Other, smaller application interests would be in SMARTs/precision medicine clinical trials and sports statistics. Methodologically, I'm probably most interested in robust model selection and record linkage, but let's not pretend I really know much of anything about what those entail. I'm interested in both academia and industry. Duke graduates people very quickly (4 years if you're going to industry or are a superstar, 5 years otherwise), and would obviously provide opportunities to branch out into application areas that aren't covered in biostatistics (e.g. time series with Mike West). I'm mostly concerned about their retirement situation -- Gelfand already retired, I think Wolpert's moving to an active emeritus position this next year, and I'm not sure how active West and Berger will be in terms of taking on advisees in the future. I know Duke is hiring several new professors this year, but I don't really know their names, so it's difficult for me to judge much of anything about their research areas or how productive they've been so far. UNC obviously narrows down the available application topics a little bit, but does allow for SMART work more so than Duke. At the same time, though, that seems to be Kosorok's area of research, and it sounds like he has a ton of students, which isn't necessarily something I'm a big fan of. UNC sounds like it graduates people roughly a year later than Duke would (~5 years industry, 5.5-6 years otherwise), but they seem to have more professors than Duke, and also seem like they aren't in the same position as Duke in terms of having several potential retirements. UNC may also provide a better statistical genetics group than Duke, headlined by Danyu Lin, but I don't think Duke's weak on that area (with Dunson, for instance). I'd welcome any opinions on which program will have more strong research advisers going forward, as well as any other opinions on which program would be a better choice. I'll note that I'm aware it's usually considered easier to go from a stat PhD to a biostatistics professorship than vice versa, but UNC does also (maybe?) require more measure theory than Duke, so I'm not clear how much that's the case here.

I've been applying directly out of undergrad -- for what it's worth, I didn't sense that the places I visited were taking mainly master's students, so that may just be an NYU preference. Regarding real analysis, I'm guessing you're going to need it for any sorts of methods development, although I've been given to understand that generally you won't be using a lot of the measure theory you learn (if you learn it -- most biostatistics departments don't really seem to emphasize measure theory, UNC and Washington excepted). I know some people do go into PhD programs without real analysis, though -- I recall someone getting into NCSU stat without real analysis a couple of years ago, and someone I met at Illinois stat's visit day said they didn't take real analysis before coming there either. If you don't take it, then you're still gonna have to cover some of it, and you'll probably be playing catch-up, so I would think it would be to your benefit to go ahead and learn real analysis if you can. Again, I haven't even taken a grad school class yet, it would be to your advantage to get opinions from the older people on here, or perhaps some Columbia statistics professors if you know any of them.

I'm a current applicant, so don't take my opinions to heart. That said, I don't think it would be terribly difficult to increase your GRE scores substantially, and that would already be very helpful to you -- I'm not sure that many PhD programs would take someone with a sub-160 Q GRE score seriously. Continued math would obviously help too. I'm a bit confused as to where you are in terms of math content, since you seem to suggest you learned probability and statistical inference before calc 3, which seems counterintuitive if not inherently impossible. That said, if you can take real analysis and do well, that would be good. If you have to take a sequences and series class first, then that won't help as much, but it still helps. Any other proofs-based classes would be helpful as well, provided you can perform well in them. I feel like you could take a couple theoretical math classes, work on your GRE studying, and reapply next year without having blown a lot of money on an unnecessary master's degree. I'm assuming these master's programs are 2 years, as well, and most PhD programs I've seen seem to only shorten by 1 year for those with an incoming master's, so I don't even think that a master's would help you finish your PhD faster in this case. Best of luck, and I hope this has been solid advice.

I probably know less than the above poster, but of your listed biostat programs, both Duke and Vanderbilt are (to my memory) both very new and very small. If you're really a fan of what those programs do, then I'm certainly not going to stop you, but you'd probably be able to find schools that are both more prestigious/potentially better (depending on research area) and easier for you to get into. I don't know if UNC biostat is too much of a reach or not, but I might suggest trying places like Minnesota biostat or UCLA biostat over one or both of those smaller programs. You said you did image data analysis, so if you're interested in medical imaging via biostatistics, you may want to consider UPenn biostatistics (in Perelman). Did you have some particular research interests/geographic preferences/etc. you'd like to bring up? I'm not quite clear what you like or don't like about these schools, except maybe some hope to be in the Research Triangle.

Brown biostat is very small, so if you're pretty interested in them then go ahead, but the small class size will make it difficult to get in.

Wow, that's a quick turnaround! Congratulations!

I mean, like always, it depends on the program. I might call Pitt biostat pretty soon, since they sent out some decisions months ago and don't appear to usually send out March decisions, but Columbia stat sent out acceptances and waitlistings recently enough that I wouldn't bother calling (and I also can pretty safely assume a rejection anyways).

Just because I'm a contrarian, I'll say that Nate's presentation at the Ohio State visit's poster session was on spatial statistics, and that the program is generally pretty Bayesian. I definitely get being overwhelmed by Ohio State's sheer size (and Columbus is pretty big too), but I guess I'd be a little more positive about their research fit for you than you are, in case you are interested in staying in Ohio. I have way different interests, though, so I'm guessing your concerns are more on the level of individual professors' research than my super blanket-y assessment. As you've likely seen, Stat PhD Now Postdoc seems pretty positive about Xiaofeng Shao's research, and looking at where he's publishing I really have to agree that he seems great. I'm having a hard time doing Iowa State justice from a cursory google search, but it seems like Philip Dixon does some spatial statistics, and spatiotemporal statistics might be a good fit for working with a visualization-focused professor like Heike Hofmann (these both had AAS publications that I saw, but they don't quite have the JASA/JRSS-B profile Shao has). Something that might be useful: how much smaller of a college town would you like? Ames is 60-70K, Urbana-Champaign comes out closer to 200K by urban area (although I have no idea how dense it is in the middle). If you wanted to go somewhere bigger, Des Moines is 600K or so and fairly close to Ames, while Chicago's more like a 2/2.5 hour drive from UIUC but is obviously gigantic (plus Indianapolis is also a 2 hour drive). What are you looking for? If you have interests that might involve national tours (e.g. musicals, major comedy shows, attending major professional sports), then I'd suggest that Illinois would be a better fit, but it's hard to prescribe that. In fact, it might help to know what factors beyond sheer research prominence are most important to you -- do these programs have similar times to degree? Does one have a really tough qualifying exam that you might worry about?

I'm guessing you're mainly looking to work with Dr. Kolaczyk at BU? He seems like a pretty good fit for your interests, and he's got recent publications in JASA and AAS (and the Journal of Machine Learning Research, which I think is also pretty good?), so I think you'd probably be in OK shape if you were working with him. I'd generally be concerned about trying to go into academia from an unranked program, but I'm guessing our more experienced posters would confirm that Dr. Kolaczyk has a sufficient background and publishing record to get you into a decent postdoc if you excelled. That said, Minnesota may necessitate more driving to get to things you're interested in, but I do kinda doubt that you wouldn't find things you'd be interested in there -- you do have a 3 million person metro area to work with, after all. If Minnesota's visit day hasn't happened yet, you may want to go and ask the students about whatever your interests are. I'm sorry for your friend, but unless y'all share a lot of interests that require warm weather and salt water, then I don't think you'd end up missing out on too much if you go there.

I'm trying to assess potential research mentors at some of the schools I'm interested in, and particularly how competitive I would be for top postdocs (and eventually, fairly strong faculty jobs) if I excelled with them (I don't have a good benchmark for "excelled", but let's say it's at a level of some of their best 5 or 10 students [depending on professor age], but not necessarily at the level of their best student). Let's say, for argument's sake, that I really wanted my first tenure-track job to be professor at a statistics or biostatistics department ranked in the USNews top 60 (combined). To achieve this, would you want your advisor to have, say, 1 publication per year in JASA/JRSS-B/AAS/JRSS-C/Biometrika (similar level conference proceedings)? How about 2? 3? If h-index would even be useful (which I'm dubious of, particularly for people who consult on medical papers a lot) what would be a reasonable value over the past 5 years -- 15? 20? If they are indeed a biostatistician and collaborate on a lot of medical papers, should that be more like 30, if any benchmark is useful? Perhaps you have a favorite pet metric from the Publish or Perish software package, or else something better to look at (say, number of active grants/active NSF or NIH grants)? How would these benchmarks increase if you wanted to teach at, say, a statistics department in the top couple tiers (e.g. Stanford through Duke, UPenn, and Columbia)? A biostatistics department in the top couple tiers (e.g. Harvard/JHU/UDub through Berkeley and Minnesota)? If this question is just way too abstract and devoid of context to be able to answer, I'll note that I'm really asking this so I can figure out for myself how "deep" Duke statistics, UNC biostatistics, and Wisconsin statistics are in terms of their numbers of pretty strong and really strong advisors, respectively, so while I'll obviously entertain literally any response to this rambling monstrosity of a topic, I'd especially appreciate any speaking to that.

I recall that Berkeley's masters program is specifically designed to be very applied and business-focused. That is generally a good fit for your goals, but you seem to suggest that you want to be doing some research, and I'm not clear that you'll really be involved in that there -- does it get mentioned on the website as being an option at all? If your interest in research did eventually make you interested in a Ph.D., I do think (based on my recollection of Cal's masters being quite applied) that Duke is a better fit. I was at Duke's visit this past week, and I recall that at least one current master's student was there as a Ph.D. program acceptance -- plus, I've heard that Duke often tries to take a couple students from its master's program into its Ph.D. program. I am not an expert, so I can't give any formal conclusions, but location-wise, I wonder whether the Triangle might even be better than Silicon Valley for biotechnology specifically. I would hazard a more confident guess, though, that Durham is more affordable than Berkeley/the bay area. If you're looking for a (pretty biased) vote, I vote for Duke, but as I hope I've noted clearly, I don't really know that much about anything relevant here, so listen to the more experienced posters preferentially.