Jump to content

Stat PhD Now Postdoc

  • Content count

  • Joined

  • Last visited

1 Follower

About Stat PhD Now Postdoc

  • Rank

Profile Information

  • Gender
    Not Telling
  • Application Season
    Not Applicable
  • Program

Recent Profile Visitors

4,714 profile views
  1. Do you know if your "smallish" Canadian school has a well-regarded reputation in the U.S.? Your profile looks pretty strong, even without the math subject GRE, but your list of schools is a bit top-heavy. I would recommend adding a few schools like Purdue, Iowa State, or NCSU, if you have the funds. Also agreed with the above, a 70th percentile should be okay. In addition, many schools don't care about the math subject GRE, and I would only advise taking it if there is a major deficiency in the application (like relatively low math GPA) or if it's required as part of the application (only a select few schools). Duke even explicitly states on their PhD Program FAQs page that they do not consider the subject GRE scores.
  2. Stat PhD Now Postdoc

    Upper Division Course Description List

    True, but if we are talking about students entering PhD programs in Stat in the U.S. who only have Bachelor's degrees, the average international student will have had more preparation than the average domestic student. They do have somewhat of an advantage in the coursework phase. But none of this matters that much as long as the PhD qualifying exams are passed, and breadth of mathematical preparation does not confer any special advantages when it comes to research (as opposed to classes/exams).
  3. Stat PhD Now Postdoc

    Low Cumulative GPA

    The cumulative GPA is a bit lower, but your math GPA and your GRE subject test score are superb. And you went to Princeton which has the best mathematics department in the country. I think you are in good shape to get admitted to a top Statistics program (if that's what you're interested in?)
  4. Stat PhD Now Postdoc

    Upper Division Course Description List

    Yes, it does seem like the undergraduate curriculum in other countries is more advanced than in the USA... I think most of the international students in my PhD cohort had already taken most of the required courses for my graduate program prior to enrolling (including the Casella-Berger sequence, linear models, measure theory, etc.) and were merely repeating the classes . No big deal, domestic students and international students are on equal footing once the research phase of the program starts. But international students do tend to have a bit of a "headstart" in the coursework phase, unless the domestic student already has a Masters in Statistics.
  5. Stat PhD Now Postdoc

    Upper Division Course Description List

    Yes, you should include anything that is taken above the level of Calculus I-III and a first course in Linear Algebra (which they the adcoms will be able to see on your transcript). I assume that you are an international student. Indeed, a course like Real Analysis is a lower division class in many foreign institutions (e.g. in China, many of the math students take analysis starting their freshman year and then have to take multiple semesters of it). But in the United States, real analysis is considered an upper division course. Real analysis, differential equations, discrete math/a first course in proofs, and advanced Linear Algebra (with proofs) could all be considered upper division math classes in the U.S. Any probability and statistics classes that involve Calculus should also be listed in your application as an "upper division" course. You could even list Regression and Design as upper division courses if you've taken them, even though at the undergraduate level, these types of classes do not tend to involve much higher level math (since these classes require some familiarity with matrix algebra/calculus and distributional theory to really study at a rigorous theoretical level).
  6. Stat PhD Now Postdoc

    Waitlist schools reapply? SOP?

    I'm not sure that "high-dimensional statistics" is very specific, but in general, it is a good idea to keep the statement of purpose fairly generic, UNLESS you're one of the rare candidates who has research experience that resulted in publications in statistics journals. Either way, though, I don't think the statement of purpose is the dealbreaker. There isn't much you can do about your GPA at this point, and the GRE is only a "sanity check." So assuming that those are decent (and I assume they were well above average, since you were waitlisted at top 10 programs), I would focus on what you *can* control (i.e., the letters of recommendation). Make sure those are topnotch (i.e. that your letter writers will refer to you as one of the best students they've ever taught,), make sure to provide your letter writers with a very detailed description of what you did/accomplished in their classes so they can tailor the letters for you, etc. It's fine to reapply to schools that previously waitlisted/rejected you. You may be lucky and be able to crack them the second time. But just remember to apply to a wider range of programs this time, not just the top 10 PhD programs. The top tier is difficult for even people with 3.7+ GPAs and near-perfect mathematics subject GRE scores to get into -- and based on what I've seen one such top 10 Statistics program, many of these programs seem to heavily weigh the prestige of the undergraduate institution (the domestic students I met were from MIT, Stanford, Yale, etc.).
  7. Stat PhD Now Postdoc

    Comparison between Canadian Ph.D. and US Ph.D.

    Publish in a respectable journal (or journals) and you will be in good shape to secure a top postdoc. Publish a first author paper in Biometrika, JASA, JRSS-B, or Annals of Statistics, and you might even be able to get a faculty job right after finishing the PhD. (For biostatistics, a publication in Biometrics would probably put you in very good shape too). That's what the most important thing for the academic job market is.
  8. Stat PhD Now Postdoc

    2019 statistics profile

    It seems like University of Georgia and University of Kentucky have funded MS programs for statistics: http://jlmartin.faculty.ku.edu/~jlmartin/masters.html Also, it seems like there are many more funded Masters programs in Mathematics than there are in Statistics, since there is always a need for graduate math TA's for the big Intro Pre-Calc and Calculus classes. I obtained an MS in Applied Math and it was completely funded with stipend in exchange for TA'ship. Two of the other Statistics PhD students from my PhD department also completed funded Masters degrees in math (one at Ohio State and one at Wake Forest), and one of my PhD advisor's former students did a completely funded Math MS degree at Clemson (and this person is now an Assistant Prof at a top 10 Stats program). If money is a concern, it may be worthwhile to consider MS programs in Mathematics or Applied Mathematics, where you load up on statistics classes as electives (that's what I did -- I took 4 Masters levels classes in Statistics, including the Casella & Berger sequence and a few applied classes). The student who got his Masters in Mathematics from Ohio State also took the Casella & Berger sequence, in addition to the required math courses for the MS degree, so he was well-prepared for the Statistics PhD program.
  9. I think as it stands, you can probably get admitted to a school at the level of Iowa State or Texas A&M -- possibly higher, given your pedigree and your strong performance in your Masters program. I don't think you would be at a disadvantage for a program because you had applied to it and gotten rejected previously (if anything, the adcom would take into consideration your most recent academic performance this time around, and your performance in your Masters program is very good). As others have mentioned, however, Duke, CMU, Harvard, Berkeley, etc. are just very difficult to get into to begin with (even a lot of applicants with 3.8+ GPA's get rejected from those schools too...), so there's no guarantee you can be admitted to one of those programs. It would be a good idea to apply to a wider range of schools, including schools like FSU, UT-Austin, and OSU.
  10. Stat PhD Now Postdoc

    European Statistics PhD

    Those are all very good schools overall, but I am not sure how their Statistics departments compare to each other. TBH, when it comes to the UK and Europe, I am more familiar with individual "big shot" names (e.g. van de Geer, van der Vaart, etc.) than I am with the departments themselves. It's often been said here that the PhD advisor's reputation matters more than the institution. I would have to agree. Someone with a renowned professor as their PhD advisor and one or two papers in respectable journals has a very good chance at securing a top postdoc in the U.S. Even if the advisor is not as well-known, having one paper published, accepted, or in revision at JRSS-B, JASA, AoS, or Biometrika gives you a huge advantage in the academic job market. My department extended a TT job offer to someone this past year mainly because they were the single author for an Annals paper (this person ultimately accepted a TT job offer from Purdue) -- this person was from Michigan State, if I recall correctly. TL;DR: the quality of publications and the recommendation letters from PhD advisor(s) and postdoc mentors seem to matter the most for academic employment. PhD granting institution is secondary to these two things.
  11. Stat PhD Now Postdoc

    GPA requirements

    I don't have any personal experience with schools below the tier that I mentioned, but assuming that you have acceptable GRE scores (which really only serves as a "sanity check") and decent to strong recommendation letters, I would consider Ohio State a reach school for you. I looked at the Results page for some of the schools, and it seems like you *might* have a shot at some places (e.g. for UT Austin, I see that someone was admitted to their Statistics PhD program with a 3.2 GPA). I wouldn't say that this is very common though, and we can only speculate (e.g. it's possible that this applicant did very poorly their first few years but then aced all their classes in their junior and senior year with high grades in all their math classes). Most of the acceptances are from people who report having higher GPAs. As I said, if you have the funds, you can apply to a few PhD programs, but I think that it would be best to prioritize Masters programs.
  12. Stat PhD Now Postdoc

    GPA requirements

    Your major GPA is a bit lower than the most competitive applicants. If your GPA in your math/stat classes were ~3.8 or higher, then your chances would have been better, even with the 3.5 overall GPA. With a 3.5 GPA in math classes, that suggests to me that you had about an even spread of A's and B's in math classes, and most competitive candidates would have more A's than B's. I think your chances would dramatically improve if you completed a Masters degree first and excelled there. With a Masters degree and strong performance there, you could probably get into one of the PhD programs in the 20-40 range for Stats (not the pooled USNWR ranking but the separated list here). If I were you and I were really determined to get a PhD in Statistics, I would prioritize Masters degrees, and maybe send out a few PhD applications if you can afford it. Then after getting the Masters, I would still target mainly schools in the 20-40 range, but apply to a few more highly ranked ones and see if you get lucky.
  13. Stat PhD Now Postdoc

    GPA requirements

    There are several factors that could be mitigating: - the prestige of your undergraduate institution and its reputation for grade deflation/academic rigor (if any). If you went to Caltech, MIT, or UChicago, etc., a GPA under 3.7 is likely to be more acceptable than a 4.0 from a regional state school. IIRC, one participant on this forum attended UChicago and got admitted to Berkeley Stats PhD with an overall GPA under 3.7 (but a GPA that was higher in the math classes). - your GPA in your major (whether or not that's higher than your overall GPA and what your major is), your GPA in math classes, and if your academic performance shows noticeable improvement over time. If your GPA is a 3.51 because of subpar performance your freshman year, that's one thing, and you can have one of your LOR writers emphasize the upward trend. If your major was physics or engineering or something and your major GPA was pretty good overall, you might still have a chance. Nevertheless, you might be out of the running for the very top tier PhD programs (if you want to do a PhD). But I don't think you're automatically out of the discussion for respectable mid-tier programs. If you could contextualize your profile better, we can give more targeted advice here.
  14. No, I wouldn't bother. That score is fine, and it isn't so low that it would raise any red flags. Sometimes if you had a lighter math background, went to a relatively unknown school, or didn't get the best math grades, taking and scoring well on the Mathematics *Subject* GRE can help boost your application. But unless one of those things applies or unless you're aiming for one of the few programs that actually requires the Subject GRE, I wouldn't recommend even taking that.

Important Information

By using this site, you agree to our Terms of Use and Privacy Policy.