Stat Assistant Professor

And even after a journal article gets accepted, it might take another year to get it published in a volume (due to the backlog of other accepted articles before it that haven't yet been published). So for an Annals of Statistics, JASA, or JRSS-B paper, you could be looking at three years from initial submission to publication. That's why some faculty opt to publish some of their work in conferences -- the review times and revision times are all in a fixed time window before the actual conference, and then the proceedings are published shortly after that. Journals are often a bit more thorough than conferences IMO (though nobody denies the quality work in some of these conferences, journal articles go through a much lengthier process, so they often contain more simulation studies and empirical analyses, more theorems, more exhaustive treatment of the problem, etc.) So yeah, I definitely recommend checking the faculty webpages, CVs, Google Scholar, and arXiv (https://arxiv.org/multi?group=grp_stat&%2Ffind=Search) to make sure that they are being reasonably productive even if they don't have a lot of new papers in press/published on their CV. It's not super uncommon for there to be a year with no publications followed by a year with like, 5-6 papers (the faculty who have multiple papers *every* year likely have a ton of collaborators, postdocs, and PhD students working with them). So definitely check to make sure the faculty are at least revising their work and putting out new work (i.e. preprints).

I would look more into it. Some faculty do not update their webpages very frequently, so they may have not just updated their website and/or CV. If they have a Google Scholar page, then that might be a good place to check to make sure that they are still being reasonably productive. For top journals, the publication pace can be a bit slow -- partly because there are often multiple rounds of revisions (e.g starting with "reject with encouragement to resubmit," then "major revision" if the first revision was satisfactory, etc.), and also because they give the authors one year to submit/resubmit. Given this, I wouldn't consider one -- or even two -- years without any publications in press/published to be a huge red flag, as long as the work is "in revision" for quality journals. However, if it is more than 2 years with *nothing* (no preprints or new papers in press), then that is potentially concerning. However, before jumping to conclusions, I would check to make sure that this is really the case, and not just the faculty member failing to keep their personal website and publicly available CV up-to-date.

Another metric to assess this is the acceptance rate of the journal/conference. The most prestigious venues will have lower acceptance rates. That's partly why the folks who publish a lot in conferences list the acceptance rate on their CV -- to indicate how highly selective it was.

I think that's pretty much it. Reputation of journal/conference, impact factor, and number of citations are all the "standard" metrics. So besides the 'top' journals (AoS, JASA, Biometrika, JRSS-B), you'll also see some high-quality work in places like Statistical Science (e.g. the original paper on GAMs appeared in Statistical Science), Biometrics, Biostatistics, Statistica Sinica, Annals of Applied Statistics, Technometrics, Scandinavian Journal of Statistics, Journal of Computational and Graphical Statistics, Bayesian Analysis, etc.

I don't think they're necessarily directly comparable, since Annals of Statistics pertains mainly to mathematical statistical theory (and is indeed the most prestigious stats journal for statistics theory). I would say among methodologists/theoreticians, Annals of Statistics is considered more prestigious. However, Annals of Applied Statistics is considered a top-tier journal and has had some very influential papers appear in it. For example, the original Bayesian additive regression trees (BART) paper (BART is one of the top-performing ML methods for prediction) and the pathwise coordinate optimization paper by Friedman et al. appeared in Annals of Applied Statistics.

University of Wisconsin seems to do better in terms of academic placements. They've had some very solid grads, like Ming Yuan at Columbia University. I also think that Madison is a better location than State College personally. If I were you, I would look carefully through faculty websites and see what journals and/or conferences the faculty are publishing in. If you see a lot of JASA, Annals (any of the stats or probability journals), JRSS, Biometrika, ICML, NeurIPS, COLT, etc., then I would consider that program to be very strong and promising for a future academic career. I would also consider things like coursework, qualifying exams, etc. At some schools, they have a lot more course requirements, and some have two written exams rather than one. That could add to the length of study. I know of some people who graduated from Wisconsin Statistics with their PhD in three or four years, so I'm not sure if they have fewer exams/course requirements.

No worries, not insulted at all. Nobody denies that the "top" programs have more famous faculty and/or faculty who are consistently publishing in top journals. Therefore, your chances of getting an academic job may be positively correlated with program ranking. However, that is only one factor; it's really on you and your track record. If you didn't attend a "top" university for your PhD, you can partly compensate for that with a prestigious postdoc, letters from famous people in the field who are familiar with your work, etc. Only you can decide for yourself if it is worth it to reapply again next year. It can be very costly and time-consuming to reapply, but the payoff could be greater if you can get better results. I think the most crucial things to consider is: if you reapply again next year, will you be a *much* more competitive applicant? And what can you do to significantly bolster your application in one year's time that you haven't already done? (e.g. can you get a higher GRE score, better recommendation letters, more research experience, etc.?) Most PhD programs in Stats don't care that much about the Math Subject GRE, and a couple points higher on the GRE probably won't make or break your application. If there's not much that you can do (e.g. because you are an international student and did not attend a "top" university in your home country), then I would just take one of the offers you do presently have.

No, I don't think you would be "crazy" for seriously considering UT Austin. UT Austin is an excellent program with great faculty, and it's located in a highly desirable location (Austin is def one of the top cities in the USA, imo). One of their PhD graduates got a TT job at UCLA Statistics.

I should qualify as well that if you're aiming to get a job at (say) Stanford or Harvard or one of those very elite schools, then your chances of doing that coming from a "lower" ranked program are probably slim, unless you're seriously amazing (very productive, tons of top publications, etc.). However, pedigree should not preclude you from getting an academic job at a fairly good school nonetheless. Even the vast majority of PhD graduates from the "elite" schools who go into academia will end up at flagship and public universities (there are only so many jobs at the "elite" schools, after all).

How "low" are you talking? Fwiw, I went to a PhD program ranked ~40 in USNWR, and we have placed PhD grads in TT faculty positions at Duke, University of Minnesota, UT Austin, etc. And I have also seen people who got their PhDs from Baylor, University of Cincinnati, and University of Illinois at Chicago (*not* UIUC) get TT jobs at Texas A&M, University of Florida, and Iowa State. It's not *just* about where you get your PhD. For example, Dave Dunson has a PhD from Emory (a very solid biostats program but not a Stanford/Berkeley/Harvard), and Michael I. Jordan (considered one of the top researchers in statistics/ML) has a PhD in cognitive science from UCSD. Both of these guys are extremely renowned in the statistics field. I can also think of other outstanding researchers who don't have PhDs from "top" schools who have done quite well in academia. I don't want to dismiss rankings completely, but pedigree really is only one factor (and byfar not the most important one). Hiring committees *really* care about your past publication record, your future potential, your postdoc experience (a very productive postdoc at a prestigious institution can help you a lot), your letters of recommendation, your PhD advisor and influential scholars who can vouch for you, your teaching experience, etc. These are all things that are taken into account for academic hiring.

Yeah, Harvard is really, really strong in the areas of causal inference and MCMC. For deep/maching learning and probability theory, I would say that Columbia, UC Berkeley, and UPenn Wharton have an edge over Harvard (e.g. you've got David Blei at Columbia, Michael Jordan and Martin Wainwright at Berkeley, Edgar Dobriban and Weijie Su at UPenn, etc.). There is also a large group of probability theory researchers in the Statistics Department at UCB, which is somewhat unusual nowadays (typically there is only one or two faculty in a Stats department working on pure probability theory topics).

I think working with an Assistant Prof is probably fine. I have seen some TT faculty who had Assistant Professors as their PhD supervisors and who still landed many campus interviews for tenure-track positions. The most important things to consider when working with an Assistant Professor are: whether their research is a good "fit" and whether they can help you to be competitive in the job market for academia or industry (either because they can help you publish in the top tier journals/conferences or because they have solid industry connections), and whether they are productive enough (by your department's standards) to earn tenure. If both criteria apply, then I say go for it. Besides, getting a TT position is the sum of many different parts, not just one thing. If your research is in a "hot" area that a hiring department currently lacks expertise in or if their job ad expresses special interest in recruiting applicants from your subfield, then I would think that you would enjoy certain advantages, regardless of who your advisor is. I also think that adcoms consider the strength of the recommendation letters too, not just whom they're written by. It is a good idea to try to get your work noticed, though, so you can hopefully get a letter of recommendation from somebody who is influential in the field. One of my letter writers when I was on the market was from a pretty prominent name in the field, and this person was neither my PhD or postdoc supervisors... but I interacted with this person fairly regularly and they were familiar with my work, so they were able to write a very good letter for me. I believe that helped a lot.

I'm sorry to hear that you are dealing with this. I don't really have much advice on how to rectify the specific issues with your department, but I do want to make a few observations. 1) It is true that international students typically have more extensive math backgrounds and are thus better prepared for the rigors of PhD coursework in statistics (e.g., it's commonly the case that a lot of international students have already taken classes at the level of Casella & Berger mathematical statistics, measure theory, etc., so in some sense, they already know the material in first-year courses). However, the gap between international and domestic students tends to narrow considerably by the third year, sometimes by the second. And by the time you start research, the majority of students are going to start out at the same level (i.e. not really knowing what they're doing). 2) If you make it past coursework and quals, then it's really the research that matters. This is what determines if you can earn your PhD -- and if you opt to stay in academia, this is what you will be judged on, not whether you earned an A or B in a core class (and if you're interested in teaching as LACs/regional comprehensives, then they will also judge your ability to teach and engage with undergraduate students). It is not unheard of for top-performing students in classes to struggle with research and take longer to finish, or for students who barely made it through quals to find their groove and excel at research. I've seen that firsthand at my own PhD institution where somebody who won "Outstanding First Year Student" struggled immensely with research and took a long time to finally finish. And other students who failed quals twice (failed first year exam once and failed PhD qualifying exam once) were still able to finish -- and even landed a TT faculty position later. 3) Those of us in academia have all failed. Even if we didn't fail classes, we probably got papers rejected, grant proposals rejected, turned down for postdocs and faculty positions we applied to, etc. So if you're 'struggling' and faced with failure, you're definitely not alone. I hope that you are able to resolve your difficulties. It is a tough situation to be in, and I am not really sure how to resolve it. Just know that if you can manage to get through the coursework, it's not all hopeless.

Agreed, UCSC is a very good program, and they have decent academic placements (if you're interested in that). They've placed some PhD graduates at University of Chicago (Matt Taddy), University of Florida, and other good places in the past. The Statistics Department is relatively new (it was part of the Applied Math department until around 2019), which is why UCSC may not be ranked in the USNWR. If you aren't interested in Bayesian statistics *at all* though, then you probably shouldn't apply there. One thing I would note though... if you decide to go into industry, I'm not sure how much of your PhD dissertation research you would really use for the "typical" jobs anyway (regardless of whether you study Bayesian or frequentist stats for your dissertation). Unless it's a very research-oriented industry job like Microsoft Research, Google Brain, or something of that sort, you probably will not use a ton of the stuff you learned in your research.

I think there are a few MS programs in Statistics that are truly competitive (in the U.S.A.)... Stanford, Yale, and Duke seem to have small Masters cohorts and are fairly selective. I would say that this is the exception rather than the rule. Even at some very elite institutions like University of Chicago and Columbia, it is not hard to get admitted to their Statistics MS program. Now, with the pandemic leading to so much virtual learning, I anticipate that schools will expand offerings for completely online Statistics MS programs, so there is even less need to be very selective about cohort size for Masters students.

I don't think domestic students are at any disadvantage for Masters programs in Statistics (and for *PhD* programs in Stat/Biostat, being domestic is actually an advantage if anything, because it's a little bit less competitive vs. for international students, NIH trainee grants can only go to U.S. citizens/permanent residents, etc.). MS programs in Stat aren't typically funded so they will tend to admit most people -- international OR domestic -- who meet the minimum program requirements for GPA, GRE Q score, and coursework (usually just Calculus I-III and Linear Algebra). I think it's just that more international students are interested in pursuing advanced degrees in Statistics (similarly with other fields like Computer Science). For that reason, you'll see more international students in most Statistics grad programs.

Yes, you should feel free to take some time off during the summers. It's not good for your mental health (nor productive, tbh) to work all the time without a break! At my PhD program, the summer was divided into two sessions (so summer classes were offered in two 6 week sessions from May to mid-June and from late June-early August). Most of the PhD students went back home (international students often went back to their home country) and/or traveled within the U.S. during one summer session. And the other session, they TA'd, RA'd, or in some cases, taught their own class.

It depends when the qualifying exams are. Some schools have them in August, in which case, you would probably spend the summer preparing. Others have them in May, so in that case, you could be done with them after May. I would say: Summer after year 1: study for quals if the exams are in August, TA or teach summer class Summer after year 2: study for quals if there is a second written exam. If not, then TA or teach, start research Summers 3 and beyond: TA or teach, continue PhD research Summer before intended graduation: summer internship I have found that many PhD students like to do a summer internship the summer before their intended graduation (some do more than one internship, but one seems to be sufficient to get your foot in the door).

If you are interested in academia, then I don't see any downside in attending UPenn Biostatistics. Somebody who graduated from UPenn Biostat this past spring got a job as Assistant Professor at Harvard Biostatistics (without having to do a postdoc). In the past, Penn Biostat has also placed someone as an Assistant Professor at Carnegie Mellon Statistics and Data Science. It depends on your publication record and your work. The two folks I alluded to were doing research in "hot" fields, which also helps too.

I am not sure what "Applicable Algebra" is. Is this a proof-based linear algebra class or abstract algebra? I think analysis and maybe advanced linear algebra should be sufficient for the Biostat programs on your list. I think you are in good shape for Biostatistics programs. Some of the top Statistics programs (Chicago, Columbia, Berkeley) may have a preference for students who have deeper math backgrounds, so it's hard to gouge your chances to the stats programs on your list -- on the other hand, you have a strong pedigree and your research experience is solid, so I'm not sure how much adcoms will take that into account. If I were you, I would apply to more top 20 schools for Statistics, like University of Minnesota, NCSU, TAMU. If you haven't taken real analysis or abstract algebra before, then I don't anticipate that you would be able to score high on the Math Subject GRE. And even those who have taken those classes need to study a lot to do well on the Subject test. Therefore, it may not be worth your time and effort. Doing well on this test may help for Columbia, UChicago, and Yale, but as far as I know, it isn't required at any school except for Stanford. And a high score on the subject test isn't really a substitute for having taken the classes and getting good grades in them.

Since you're performing very well academically at an Ivy League School known for slight grade deflation and you have great research experience with some papers in preparation, I think your chances of getting into a top program are quite good. I think you should be able to get into those Biostatistics programs. For Stat, I don't think you need to apply to Northwestern. If you want to add a few "safer" options, I would add Texas A&M, UCLA, and NCSU (these aren't "safe" schools in general, but specifically for your profile, I think that they are very safe bets). That said, it is worth noting that Columbia Biostatistics is not in the same league as Harvard, UW, JHU, or UNC for BIostat. And some of the programs you have listed are quite different. For instance, Yale S&DS strikes me as quite theoretical, while Columbia Biostat is quite applied. You may want to take stuff like that into account. If you want to do less theoretical stats, then Berkeley Statistics is a good choice (UCB is strong all-around in probability theory, thoeretical stats, and applied stats), and I would also add some schools like University of Washington (they have great faculty in demography and social science stats, for example).

The most useful math classes to take, IMO, would be advanced linear algebra, real analysis, and optimization. Is Advanced Calculus I-II the undergrad sequence in real analysis? If so, then I don't think you need to bother with "Applied Real Analysis." Since you're in a Statistics MS program, I think just taking statistics classes, plus analysis, advanced linear algebra, and one or two other advanced undergrad math classes should suffice. In your case, I think your long reaches are all unrealistic. I would not spend money applying to those programs. I think your list of "reaches" should also be trimmed down, and you should target programs like University of Missouri, University of South Carolina, Kansas State, University of Maryland Baltimore County, etc. Since you seem like a somewhat nontraditional student, your statement of purpose and letters of recommendation might also carry more weight. So I would definitely spend some time crafting your SOP and detailing your motivations for wanting to get a Statistics PhD, your mathematical preparation, etc.

I think it depends more on what field you're applying uncertainty quantification to rather than broadly "uncertainty quantification." I have seen some work published in top journals related to uncertainty quantification for deep learning and causal inference. A big research area of Bayesian nonparametrics right now is studying the coverage properties of posterior credible sets (i.e. under what conditions a credible set is also a confidence set that gives the same asymptotic coverage). It seems like inference and uncertainty quantification are in general always seen as more important by statisticians than people in Computer Science/machine learning departments. That's why there are a lot more articles in statistics journals than machine learning conferences related to post-selection inference, construction of simultaneous confidence intervals, etc. Getting hired in academia (at research universities) is highly dependent on your publication record and your letters of recommendations. The former depends a lot on the novelty of your work. It is possible for somebody whose work is on time series to get hired, but their contributions to the area have to be novel. FWIW, it seems like multivariate time series (such as vector autoregressive models) is still a fairly "hot" topic, and I know of somebody who was hired as an Assistant Professor at Cornell Statistics because of an Annals of Statistics paper that was a significant contribution to regularized VAR models.

There are definitely some people who hold dual MD and PhD's who work in computational biology and/or bioinformatics, and they have established labs in medical schools for this. I am not sure how common it is for someone with only an MD but not a PhD to work in this area, though.

You could check out biostatistics PhD programs. That might be a better fit if you're certain that you are industry-oriented and you're more interested in applications than methods/theory. If you post your full profile (whether you're a domestic or international student, your overall GPA and your GPA in math classes, math classes taken and grades received in those classes, an idea of the college you attend -- e.g. range of rankings in USNWR, etc.), we can give you an idea about how competitive you are.