cyclooxygenase

Greetings everyone! I'll defer any commentary about Duke's coursework to another recent post of mine, but I wanted to clarify a couple things: 1. I think(?) Duke got referred as having 2 qualifying exams, but we only have 1! We have the actual, appropriately named Qualifying Exam (formerly the First Year Exam -- it's still administered after the first year), and then a later Preliminary Exam, which I'm guessing is what's getting confused for a second qual. I'm not entirely sure how to characterize the Prelim Exam myself (the description of it is here), but I'd probably call it closer to a thesis proposal than a qualifying exam in the usual sense, at least functionally. 2. (Sigh...) I've lived in Morningside Heights before, I've lived in Durham before. Anecdotally, I've been fine both places (and I assume I would've been in Boston as well). Regardless of that, I'm really not convinced that debating the safety of large, highly-populous neighborhoods or entire cities without some rigorous framework is a productive conversation, y'all.

Speaking specifically about Duke: Stochastic processes (specifically applied to MCMC and HMMs and such) get covered as a big part of a second-semester class ("Probability and Statistical Models"). That said, how important stochastic processes are in that class varies by Prof -- if Mike West teaches it then you'll probably eat autoregressive models and stochastic processes for breakfast (and he's a wonderful teacher, so you'll enjoy it!), but if a different prof teaches that class then you might get a bit less stochastic processes and s bit more E-M alg, finite state space MC, and so on. My point being, what you need to know (at least for the qualifying exam) will be taught, they're not just going to assume you've learned stochastic processes ahead of time. Martingales are covered in the measure theory course most of the time, although since they're at the end they're liable to get abbreviated if stuff runs over. Again, you'll learn what you need, at least for quals. Real analysis is definitely helpful in prepping for the measure theory course, though. I will note that it might be possible for you to skip measure theory and the intro Bayesian course (mentioned below) if you have strong background there already, although I think that's less likely (/impossible?) to be the case in the future than it has been in the past. Duke has you take a Bayesian class first semester (this is the undergrad/master version, but an OK idea of the topics anyways), but even so it's nice to have basic background in conjugate prior sorts of models and coding basic Gibbs samplers by hand (you know what I mean, no JAGS or Stan). If you have that you're likely fine, although the intro Bayesian class is switching from combined MS/PhD to having a PhD-specific class that may cover more advanced topics going forward. I'd definitely recommend going over Casella & Berger if you're rusty, since the inference class (which is second semester) is probably the most important class for quals and is taught from TPE2, so there's some assumption you're familiar with C&B and maybe CMT and Slutsky's theorem that aren't covered all that much in the inference class (or they come up right at the end) but that will be relevant on the quals + potentially going forward. Code-wise, it's mostly R in classes, although some profs prefer MATLAB. For topics to learn for research, Stat Asst. Prof's your person, as you already realized.

Like Stat Assistant Professor says, there are a couple Duke connections that do this sort of stuff, but they both have PhD's. Hau-Tieng Wu (https://hautiengwu.wordpress.com/home/) got his MD first, then did an applied math PhD at Princeton after practicing I believe, while Justin Silverman (http://www.justin-silverman.com/) did his MD and PhD at the same time.

Obviously I'm pretty well debunked as a good evaluator, but just for purpose of giving you some opinion to ruminate on: The health problem background helps limit the effect your major and overall GPA will have on admissions. While a higher quantitative GRE score wouldn't hurt, you have overall good GRE scores, pretty strong math grades, and aren't completely lacking in research experience. I think you'd be competitive most anywhere -- obviously the top schools (Stanford, UChicago, etc?) are never really a sure bet, so have some other schools in mind, but I think you'd have at least a shot. I mean, I may be applying in Biostat, but feel free to look at my evaluation thread: https://forum.thegradcafe.com/topic/110483-2019-statbiostat-phd-applicant-judgment/?tab=comments#comment-1058618664. Without the F's we'd probably have reasonably similar GPAs, my quantitative GRE is better but yours really isn't a big worry so far as I know, and our research experience is fairly comparable. Meanwhile, you absolutely blow me away in math background. While I don't have much shot at applying to the sorts of top institutions I'm naming to you (especially in statistics, rather than biostat), I'm applying for Ph.D. programs rather than M.S. programs, and I feel like I might actually have a shot at some top programs if I had your math background. Hence, I figure you would do better than I would if I was applying to M.S. programs, and I would expect to get into a fairly decent M.S. program, if not quite the level I'm suggesting you could apply for. That's my reasoning, take it for what little it's worth.

You haven't included much in the way of grades by class, which might be more useful. For instance, what were your grades in Real Analysis (which I'm assuming you just called Analysis) and your Measure Theory class? Those would be most important. That being said, if you got A's in both of those, I really have to assume you would be very competitive for most Master's programs, that's a pretty strong math background you've got there, probably more than most people they run into in Master's admissions.

Your background in real analysis, measure theory, and to a certain extent topology is definitely preferred. You already took probability and did well, with your theoretical math background I doubt anyone would be too concerned about your ability to learn statistical inference. I could see the very top places be concerned with A-minuses, but in general you've done about everything mathematically a program would want, at least from what the forums here seem to say. I'd generally agree with statfan about being competitive for decent Ph.D. programs -- it's a little hard for me to judge difficulty of admissions in Stat vs. my aspirant Biostat, but you've got a substantially stronger math background than I do with similar research, you should be OK regardless of what applications you choose to put out.

Your fairly strong grades in Real Analysis I & II, plus Topology and Measure Theory, make me think you'd be able to get into most any master's program, provided you have reasonable GRE scores. Since apparently schools actually care about GRE scores, I'd recommend studying real hard for them so that an admissions committee can at least point at good GRE scores to contrast with your major GPA (and maybe your overall GPA, although obviously that's less of an issue). Without strong GRE scores, I would assume some good places would take a chance on your due to your aforementioned math background, but the very top places (and some lower down) might write you off for your major GPA. Reasonably useful questions: 1. What is your major GPA without the F? 2. Did you take the GRE via computer, and if so, what were the predicted scores? They aren't likely to change much, if at all.

I'm glad to hear! I'm quite in favor of heavy theoretical training, I just find applications being involved to provide some extra motivation.

I was given to understand that UNC, certainly in statistics but potentially in biostatistics as well (hard to suss out from individuals' comments), was so given to theoretical study that Ph.D. students would spend their entire time in the program without doing anything involving real-world data. That particular anecdote may be relating the statistics/operations research department, in which case I'll agree that my comment is odd. I'd heard other reasons I wouldn't be a good fit with UNC Biostats from other people, but give that such accounts are just as anecdotal as yours (or perhaps moreso), I'll happily give it another look, so thank you for your input! That's it! As for Michigan, a cursory examination of their biostat faculty didn't turn up any pharmacogenomics research at first, hence my leaving it off, but I'll give it some more passes. Thanks guys!

Thank goodness, a response, and from another Packers fan too! The letters of recommendation are (understandably) what I'm most concerned about, so your circumspect reply is unsurprising. As for your school recommendations, thank you! I am interested in looking at data within the next 6 years, though, hence why I've left UNC off the list in particular.

I'm definitely not qualified to judge you, but I'll go ahead and chime in so that someone in the field gets annoyed enough to respond and give you an actual evaluation: If you're applying to a Master's with two linear algebra courses (not sure what to make of Intro Linear Algebra, but I'll assume for now that it's a standard linear algebra class and your forthcoming Linear Algebra class is more advanced), Real Analysis, a putative 4.0 GPA and a Math major's worth of proofs background, then I'd hazard a guess that you could stand a reasonable chance of admission for most any Master's program, including those you listed, provided the work experience you're getting post-baccalaureate is statistics-related (which, given your internships, I'd assume so). I'm not sure how Master's admissions committees view research experience, but your lack of that seems to be your only outright weakness, and job experience can certainly help allay such worries, even if it doesn't wholly dispel them. As for your listed concerns, just get A's in Linear Algebra and Real Analysis, and if you're concerned take an extra proofs-based class, but I think that you would be fine regardless. I doubt any department cares whether or not Math was your first major, and you can just explain Business Analytics briefly in your personal statement, or course-by-course as (maybe?) an addendum to your CV. I would ask if one of your internship advisers/superiors, or preferably your future boss (even better if they're the same person), could write a recommendation for you -- three letters from Professors, none of which you've done research with, seems a bit monotone to my untrained eye, and you'd probably want to sell your communication skills and business experience so as to make yourself appear more fleshed-out as an applicant.

It's definitely not weird to do your Ph.D. somewhere different than your Master's -- in fact, just anecdotally, it seems pretty dang common. I think you'd have to go through Ph.D. application and admissions rather than transferring, though, which seems contrary to what you suggested. I'm about the least knowledgeable person on this forum, but re: moving from a Math background to a Stat background, I'm guessing many professors would be thrilled to have applicants with your background in theoretical math, regardless of whether you have a Master's or not.

Not quite what you were looking for, but for climatology, you could also look at Atmospheric Sciences programs instead. I'm given to understand most math-heavy students in such programs go more into dynamics than climate research, but regardless -- if you wanted to look at those, off the top of my head good programs would be Colorado State, Oklahoma, Washington (Seattle), Wisconsin (although less so these days), MIT (ditto?), and so on. If you're really interested in that there are better sources than me, of course -- I'm just working off of my mom's time in climatology.

Thanks for any feedback, good luck to all other applicants! Undergrad: Large Public University in the US, Top-75 on US News but not really a science powerhouse Majors: Statistics, Neuropsychology GPA: 3.80 Programs Applying: Statistics Ph.D./Biostatistics Ph.D. (Statistical Genetics/Pharmacogenomics, hopefully) Type of Student: Domestic Caucasian Male Relevant Courses: Linear Algebra for Physics/Engineers (A), Residue Calc & Linear Algebra for Physics/Engineers (A), Integral Calc (B+), Multivariable Calculus.(A-), Diff Eq (A), Intro to Analysis (A-), Basic Proofs (A), Mathematical Biology (A), Regression Analysis (A), ANOVA/Experimental Design (A), Math Stat (A), Programming in R & SAS (A), Genetics (A-), Neural Genetics (A-), Neurobiology (A), Psychopharmacology (A-) GRE: 168V / 170Q / 3.0W Research Experience: Working with a statistician fairly recently, have some research background via psychology. Have some student presentations in psychology, weren't qualitative/nonstatistical but weren't especially intensive mathematically/computationally. Recommendation Letters: Not super strong, will probably be professors from classes who know me well enough. Research is either too recent or too far back to be super helpful for letters. Coding Background: R, SAS, and Java Applying to: Carnegie Mellon (Stat), NC State (Stat), Iowa State (Stat), Minnesota (Biostat), Wisconsin (Biostat), Yale (Biostat), UPenn (Biostat), Emory (Biostat), Duke (Biostat), Brown (Biostat) -- haven't finished researching lower-ranked options Notes: I'm taking Real Analysis and Statistical Inference this fall (Real Analysis =/= Intro to Analysis, for clarity), hopefully Measure Theory and Numerical Analysis in the Spring, along with some more genetics and biology. So, how realistic are the schools I've picked out so far? Any recommendations for schools from those who are also aspirant statistical geneticists? I'm open to suggestions of lower-ranked schools than mentioned, I just haven't finished picking those out yet.