Stat Assistant Professor

How much mathematics have you taken beyond Calculus I-III and Linear Algebra? Have you taken any proof-intensive courses, namely Real Analysis? Based on your description, I will assume that you attended a school like IIT Bombay or IIT Delhi, which should be favorably viewed by admissions committees. Most adcoms will likely either have at least one member who is from India, or if not, they will be able to consult an Indian colleague in their department about your school. So I am sure that they will be aware of the grade deflation and the rigor of your institution. I think you could potentially have a shot at some Statistics programs in the top 30 if you have taken some advanced mathematics beyond the Calculus sequence and LA. However, you would be competing against some very strong applicants from Indian Statistical Institute (all of whom either have or are in the process of obtaining Masters in Statistics and have already taken classes like measure theory). So I would consider a school like University of Michigan to be a more appropriate "reach" school for your profile (I know of some people who attended the IIT's for engineering and then went to a school like UM for a PhD in Statistics).

You have an excellent GPA in Statistics/IE from one of the best schools in South Korea. I hardly think you have "no chance at all." However, you may need to calibrate your expectations a bit (e.g. Berkeley) since admissions is extremely competitive for international students. But being a strong student from Yonsei certainly gives you a great chance at schools like NCSU, TAMU, Purdue, Penn State, etc. I would target larger programs at big public universities and apply to a few "reaches." I would just apply and see what happens. I mean... you miss 100% of the shots you don't take. Try to get very strong letters of recommendation.

Students from the top universities in South Korea, including Yonsei, tend to do pretty well in Statistics PhD admissions. If you get A's in Real Analysis I & II (I would take two semesters of it if you're worried about the lack of proof-intensive classes), that will help your profile a lot. That said, it is extremely competitive to be admitted to UC Berkeley, Stanford, etc., especially for international students, and you will be competing with a lot of international students (including from Yonsei, SNU, KAIST, etc.) who have a lot more advanced math than you. In fact, if I recall correctly, when you submit your PhD application for the Statistics program at Berkeley, you need to submit a list of ALL the math classes you have taken, the grade you earned, the material that was covered, and the textbook(s) used. So yeah... these schools will tend to favor students with heavy math backgrounds, much moreso than those who have taken a lot of applied statistics classes. As far as top 30 schools, I think you would have a very good shot at being accepted at a large state school like Texas A&M, Purdue, or Penn State. I think this may be the target range for your profile, but you can also apply to several ranked higher than that for good measure.

I think you should have an above average shot at NCSU and Duke. After all, you did go to a top liberal arts college and did pretty well in the math and stats classes you did take. Your profile is by no means a "shoe-in" at these places, but I think that if you can get excellent letters of recommendation, I wouldn't discount your chances at either of these schools. Your profile reminds me: there is an outstanding statistics researcher James Johndrow (now an Assistant Professor at Penn Wharton) who got his PhD in Statistical Science from Duke even though his undergrad degree was in Chemistry (also obtained from a top liberal arts college). And I also know of a young woman whose Bachelor's is in Psychology/Pre-med from Columbia but she also got her PhD in Statistics from NCSU (now a postdoc at Columbia). I think that Duke and NCSU are more open to accepting applicants from "lighter" math backgrounds and different majors than other places -- I know of at least a couple students/alumni from NCSU and Duke Statistics who did not have extensive math backgrounds. However, they did come from strong pedigrees, so your chances may be inversely proportional to how prestigious your undergrad is (that is, you can possibly get away with a lighter math background if your undergrad institution is very prestigious. But if you went to an unknown school, then you need to have very strong performance in math classes to be competitive). However, UPenn Wharton, CMU, and Columbia may be tough for you to break though without a lot more math, and I would recommend that you apply to a wide range of PhD programs, like the other posters have suggested. I also think you can definitely get into a top 10 Biostats PhD program, no question, if you think that this aligns better with your interests than traditional Statistics programs.

I think that earning A's in the more advanced classes can mitigate the B's in Calc III and Linear Algebra. But you may want to take more advanced math classes to demonstrate that you can get *consistent* A's in advanced upper division math courses. Is there a proof-based upper division linear algebra class you can take? If so, you could try to ace that class and that would certainly help your application if you were to apply to math/statistics PhD programs. For math PhD programs, it is my impression that many of these programs want to see that you've taken a few PhD-level courses too (e.g. measure theory or topology).

UCSD Mathematics has some great statistics faculty, particularly for high-dimensional statistics and ML. Many of them seem to be publishing in the top statistics journals and ML conferences, so you could do quite well there if you were to go to UCSD and be supervised by one of these professors. There are also a lot of great faculty in the UCSD Halıcıoğlu Data Science Institute whom you can work with as a PhD student, and that can also help set you up well for a career (academic or non-academic) later. I don't think your opportunities would be limited if you were to go there, to be honest. There are some excellent mathematics departments that house statistics within them, and I would count UCSD as one of them.

Yes, I think your MS school is good enough to apply for schools in the 30-50 range. I might suggest adding a few more schools in the range of UF (e.g. FSU, UConn) and Mizzou (e.g. UC Rirverside, Kansas State) to be safe.

WIth a 3.1 from Carnegie Mellon and that GRE score, you would probably be able to get into most Data Science MS programs, outside of a few that are truly selective/competitive. Even then, I wouldn't rule out your chances. I would look at program websites and see what their basic requirements are. I assume most of them require a GPA that is 3.0 or higher, and if you meet that threshold, you probably stand an excellent chance.

For the vast majority of Masters programs in Statistics/Biostatistics, you mostly just need to have gotten a B- or higher in Calculus I-III and Linear Algebra and have a sufficiently high GRE Quant score. You didn't post your grades in Calc I-III or LA, but I'm assuming that they are fine. You should have no difficulty getting into these Masters programs. If you are interested in getting a PhD in Statistics eventually, then take Real Analysis during your Masters program and maybe one or two other proof-based math classes.

I would suggest that you post your profile over at https://mathematicsgre.com/ They can probably give you a better assessment over there. Based on my limited expertise however... unfortunately, I do think that you are right that your undergrad GPA and GRE subject may not make the "cut-off" for some schools. On the other hand, though, you did take a lot of graduate-level math classes and did well in most of your math courses. I think other schools may be more open to your application, provided that: a) it is explained in your letters of recommendation and your statement of purpose that, apart from a health issue one semester, your math GPA is excellent and your overall/math GPA's showed a clear upward trend (i.e. "My GPA improved from [freshman year GPA] to [current GPA]"), and b) you and your letter writers emphasize that you took 10 graduate-level courses (and presumably did well in all/most of them). There is quite a bit of variation in how much different departments emphasize the math subject GRE.

Another main difference, I would say, is that it is virtually impossible to get into a CS PhD program without any research experience, whereas Stat/Biostat departments tend to de-emphasize research experience in favor of mathematical maturity. In CS, a 3.2 GPA (either cumulative or major) could be mitigated by strong research experience and a first/second-author publication in a prestigious ML conference, whereas Stat/Biostat departments place greater emphasis on grades (especially in math classes). In fact, for CS departments, the research practically *is* your application (see http://www.pgbovine.net/PhD-application-tips.htm) But yeah, I agree with @StatsG0d that statistics departments will often emphasize things like asymptotic properties, uncertainty quantification, and mathematical foundations, whereas (applied) machine learning research groups in CS departments do not seem to care as much about theoretical properties or statistical inference/uncertainty quantification. Predictive accuracy and computational efficiency are generally what is emphasized in CS. The culture is slowly changing at a lot of Stat departments though, with more and more departments placing high value on ML conference papers in top conferences rather than only mathematical papers in traditional (bio)stat journals. And nowadays, a lot of papers in JASA, JRSS-B, and Biometrika feature novel contributions to statistical computing, not just fancy-schmancy math. In the Bayesian statistics community, it is also now well-acknowledged that scalability and computational efficiency are essential concerns from practitioners that we have to address. So the lines between CS and Stat are becoming blurrier... but still, certain topics like NLP and computer vision are almost exclusively in CS.

Pacific Islander/Native Hawaiian could be considered URM, but there are also other URMs who apply to the top (Bio)statistics programs that have better pedigrees and/or stronger academic records and high GRE scores. So I don't think that fact will automatically make you a "shoe-in" at the top PhD programs. For reference, these are the types of URM's that are accepted into top PhD programs in Statistics: https://aswilson07.github.io/website/ https://vivo.brown.edu/display/lcrawfo1 https://provost.ncsu.edu/news/2017/02/statistical-success/ https://anson.ucdavis.edu/~melopes/ You will notice that they all attended top schools and/or graduated summa cum laude, etc. I would focus mainly on PhD applications to schools in the 20-40.range

Of the top 20 programs in the USNWR rankings for Statistics (including the ties), I would say you might have a chance at Iowa State and Texas A&M, but the others like Cornell and UNC-Chapel Hill (and any higher than that) will be quite challenging for you to get into, with your overall GPA. I would suggest focusing on getting all A's in your Masters program and making sure that one (or several) of your LOR writers emphasize the upward trend in your grades and the A's in Real Analysis and other upper division math classes. But make sure that it is framed in a positive way -- like "My overall GPA improved from a [freshman year GPA] to [GPA at the end of undergrad], and I have received almost all A/A-'s in upper division classes, including [real analysis, statistical inference, etc.]" I would focus on mainly schools in the 21-40 range, with a a handful of schools ranked lower than that.

Most would consider the top journals in Statistics to be: Journal of the American Statistical Association Annals of Statistics Journal of the Royal Statistical Society Biometrika Biometrics

I don't think Penn has a Statistics Masters. Do you mean their Masters in Data Science? Anyway, I think you have a solid chance at getting admitted to those programs. Just out of curiosity: is your ultimate goal to work in industry in the U.S.? If so, given the current climate, I would say that it is far easier for international students to get an H-1B work visa with a PhD in STEM (Statistics has recently been classified under this STEM umbrella) than it is for those with only a Masters. Even non-citizen/non-PR STEM PhD's from mid-tier universities will have an easier time finding industry employment than those with Masters from the top-tier universities. This is something that I would consider too. Did you take real analysis as part of your undergrad? If not, then you may want to consider taking that in your MS program to prepare yourself for PhD applications if need be.

I'd note also that many PhD students do not publish in the very top journals or very top conferences (myself included -- the publications from my PhD thesis research ultimately appeared in quality journals but not the top 4). Most of these PhD students will probably go into industry, but if you aim to stay in academia and your primary goal is to work at a research university, then you can do a postdoc to improve your profile. If you don't have any publications in the top journals as you're finishing your PhD, I would recommend doing a postdoc at the most prestigious university you can and working with PI's who have a track record of publishing in the top journals. In my case, doing a postdoc really helped a lot and made my profile much more competitive. As I was finishing my PhD, my profile wasn't competitive at all for most research universities (I did make the shortlist for an R2, more teaching-intensive university though, but ultimately did not get that position and opted to do a postdoc instead).

For example, this guy was hired by University of Michigan Statistics Department last year with a bunch of ICML and NeurIPS publications, rather than pubs in traditional mathematical statistics journals. https://regier.stat.lsa.umich.edu He has a *lot* of them though. I cannot confirm if this is indeed the case, but based on personal experience, it seems that if your pubs are mainly in journals, then 1-2 articles in top tier journals gets you on the "in discussion" shortlist for many search committees. This is typically because it takes much longer to publish a journal article... whereas with conferences, the review time and rebuttal times are on a fixed schedule (so you only have [x] weeks to reply to the referees who only had [x] weeks to peer review your paper) so they go through much faster. So it may be the case that more conference papers are required to make an impression. Is your ultimate goal to work at a research university? If you aim to go to a primarily undergrad institution, then publications in top venues is not nearly as crucial (but you do need some publications and you need to show potential to churn out some more in the future... but they need not be top-tier).

I think a lot of your chances in the academic job market will depend greatly on how "traditional" the department is in its makeup. There are some Statistics departments that are partial to hiring people whose PhDs are in Computer Science and/or whose publications appear mainly in venues like IEEE, ICML, and NeurIPS rather than the traditional (bio)stat journals like JASA, Biometrika, Annals of Statistics, JRSS-B, or Biometrics. You will need to take a look at Assistant Professors' profiles on departmental websites to see where recent hires have been publishing to see how amenable a department is to publications in ML conferences or IEEE proceedings. The academic job market is pretty tough though, and without at least one paper in a top 5 stat journal (including under invited revision) or a top tier conference (acceptance rate < 10%), it can be hard to even make the shortlist at a lot of schools. For Computer Science departments, journal papers are not nearly as prestigious as conference papers, so if you aim for CS faculty jobs, you typically need 3-5 ICML/NeurIPS papers to be a competitive candidate.

There is the country Samoa and the territory American Samoa. If he is from the former, then he's an international applicant. But if he is from American Samoa, then he is American and his chances do change a bit. Even if he were domestic, I would still say the top 20 schools are reaches, though. But he might be able to get into some schools like Iowa State and some others between Penn State and Rutgers.

Agreed with you that doing those things would help improve the OP's profile and chances. But I think top 20 is a bit optimistic since he is an international student, and there is much less room for error there given all of the competition. While I would agree with your assessment if he were a domestic student, the fact is that most international students in the top 20 (really, 30) schools will be those who graduated with high honors from the top universities in their respective home countries or in the U.S. University of Utah is a fine school, but adcoms are likely to view a top student from Peking, Tsinghua, ISI, etc. more favorably. That said, I do think that strong performance in a Masters program might make a school like University of Florida (and similarly ranked programs like UConn, FSU, etc.) attainable for OP. There are some international students I know who graduated from UF who attended undergrad institutions that were fine but not extremely prestigious (such as Bard College and University of New Hampshire). I think schools ranked 40-70 would be most appropriate for the OP's profile, but he can certainly try a few schools ranked above and below that for good measure.

I don't think you have any chance at UChicago or CMU. UCLA may also be tough. I think the top 30 Stat PhD programs (according to USNWR) may be out of your reach based on your cumulative GPA and the fact that competition among international students is very fierce. If I were you, I would start at the tier of Ohio State and University of Florida and work your way down. I think that conditional on strong performance in the first year of your Masters programs, programs ranked in the 60-80 range of USNWR may be your most realistic bet -- provided that the discrepancy between your overall GPA and your major GPA's is adequately explained in your application. It would probably be best to ask a letter of recommendation writer to point out the marked improvement and the fact that your math GPA is a lot higher, with mostly A/A's. They should also draw specific attention to the A/A-'s in Real Analysis and Statistical Inference in their letters, and you can also point these out in your statement of purpose.

The top statistics departments should have at least a few faculty who publish consistently in top ML conferences, including ICML and NeurIPS. You can check the faculty pages to see if there are any who publish in such venues. Agreed with bayessays that CS would be much better for things like NLP and computer vision though.

Of the statistics programs on your list, I think you will definitely get into MSU and Purdue. I wouldn't rule out any of the others in your top six, since a 3.7 from one of the top three schools in South Korea is very impressive. I have seen students from Indian Statistical Institute with cumulative GPAs in the low 80s (their grading scale is graded out of 100) get accepted to the likes of UPenn Wharton, Columbia, etc., and I assume that a 3.7/4 is similarly considered very good from a top school in South Korea. I think most adcoms are aware of the high level of rigor at the top math/stat programs in India, China, and South Korea, plus how advanced the coursework is (i.e. many undergraduates at these top schools take measure theory as part of their undergrad education, which is less common for domestic applicants). I am not very familiar with the math programs on your list. I would recommend you post your profile at mathematicsgre.com to see what their opinion over there is about your chances. If you are trying to narrow your list to only three or so schools from the last eight schools on your list, though, I would definitely also consider things like quality of life (e.g. whether you strongly prefer to live in a city like Seattle vs. a small town like College Station, or if you are indifferent about that). This is somewhere you will be spending 4-6 years of your life, so don't discount things like that.

For Statistics programs, I would recommend checking out Stanford, UPenn Wharton, UC Berkeley, and UNC Chapel Hill. These all have some really great faculty in probability theory. I think with your profile (lots of math courses and a strong GPA from one of the top universities in South Korea), you have a great shot for UNC-CH STOR. The others on this list are extremely competitive for international students, but you should have a decent chance. Try to get strong recommendation letters from faculty who know you well, can write that you are one of the top students, and can point to specific examples of your research potential, etc.

UPenn Wharton Department of Statistics also has some very strong probability faculty (Bhaswar Bhattacharya and Jian Ding). In particular, Bhattacharya at Penn Wharton is an expert on combinatorial probability, so that would be an excellent fit for your research interests. And Jian Ding and his PhD students regularly publish in top probability theory journals like "Annals of Probability" and "Probability Theory and Related Fields." Which Asian nation are you from, if you don't mind me asking? If you are from China, South Korea, or India, your profile is pretty strong and you stand a good chance. If you are from another country, competition will be tougher (only because there are already so many excellent applicants from the aforementioned countries). Other than that, there will probably be a few differences between Statistics and Math programs. First, most Stat programs will tend to be dominated by statisticians rather than probabilitists, though I'm sure there will be at least one other member in your PhD cohort who will share your interest in probability. Second, the coursework will be different, i.e. you wouldn't need to take classes in topology, abstract algebra, etc. in a Statistics department like you would in a Math department. However, you would still need to take a few proof-intensive, heavily theoretical classes in a Stat department like measure-theoretic probability theory, theory of linear models, advanced statistical inference/decision theory, etc. As far as getting postdocs in the field of probability (many of which will be in math departments), this shouldn't be any problem for someone coming from a Statistics Department as long as they have great recommendation letters from well-regarded professors and as long as they have one article in a top journal like Annals of Probability or Annals of Applied Probability. As far as I'm concerned, as long as there are at least some probability theory faculty in a Statistics department, you could do just as well there as you would in a math department. Most of the stuff you do for your dissertation research is stuff you have to teach yourself on your own anyway, not things from classes (though the classes will give you a basic foundation for more advanced self-study).