bayessays

An MS in biostatistics is, for all intents and purposes, the same as an MS in statistics. IUPUI also has a PhD program. It was just an example though, but there is pretty much no job that would want an MS in statistics where and MS in biostatistics would not be acceptable. The only programs that look at math subject GRE scores are a handful of elite PhD programs. Most top statistics PhD programs do not even recommend the test. If you apply to some random MS program and submit your math GRE subject score, they may not even know how to interpret it.

Agreed with above. Unless there's a person you particularly want to work with at another top department, CMU is great. Lots of great ML and applied work, as well as some just absolutely top statisticians like Wasserman, Kass, Shalizi and others. Another pro is that CMU is one of the best value programs since they give you a big stipend in a relatively cheap location.

I'd be biased towards the option that will save you money. Wake Forest is a respectable school and think the thesis option could be very helpful for PhD admissions in future. It's not in as big of a city as Minneapolis, but it's not the middle of nowhere and there are bigger cities in North Carolina within a few hours (also much better winters). I don't think the ranking is a big enough deal to be worth the extra money, especially considering you probably can form closer ties with professors at Wake Forest. The MA/MS distinction is meaningless.

This is true. But I think most schools don't want someone to attend who doesn't want to go there, so these are generally not an issue. This is just from what I've heard though.

Your linear algebra grade is a little low, but I got a B- and was fine for PhD programs and you have a pretty good analysis and some good calculus grades. I think you might be able to convince a department to give you a chance for a master's especially if you do very well on the GREQ. Have you looked at some of the programs in Indiana like this one? https://fsph.iupui.edu/admissions/apply/ms-bio-applicants.html

I don't think the subject GRE will be worth the studying effort for a master's program. There may be an unranked program you could attend that will allow you to come because you're paying your own way, but some schools have 3.0 cutoffs for admissions. How did you do in the calculus and linear algebra classes? If you could get decent grades in those and maybe a prob/stats class and a high GRE Q that would probably help. I don't think it's hopeless, but you'll have to find a department that doesn't have a cutoff that will automatically disqualify you from applying.

In that case, that is pretty impressive and your list is probably fine. I'm not super familiar with optimal transport except knowing there is a guy who works on it at UIUC which I think you'll probably get into. It will be hard to find lower ranked schools with many theoretical ML people. Maybe check out UC Davis as one more "safer" option.

I don't like to comment on international profiles because the admissions are so competitive that it's hard to even guess. Next year, there may be funding issues as well because of the virus situation so things might be even more competitive. My guess is that you might want to add a few more of those "safety" schools because I don't think they're totally safe. At visit days, I met an international student with an MS degree from a top US school who had a JMLR paper and got into only 1 school - one of the safeties on your list. If you're the first/main author on some of these papers and your letters of recommendation are great, you'll probably have really good results.

I don't agree at all with above but you have a good profile and should apply to top schools as well as plenty that are not in the top 10 and probably some that aren't in top 20.

Ah sorry yeah. I've seen departments talk about faculty hiring freezes and ways to cut costs already with the recession, so I think it is reasonable to think admissions (funding, really) might be more competitive next year - especially if students admitted this year are allowed to defer. I think it will certainly be worth it to apply unless the department says differently though, but we'll have to see what happens.

There's another thread on this topic below. Basically, nobody knows but it's possible classes might be online in the fall, people might be able to defer admissions, there might be funding issues at some departments.

Yes, I did not mean to imply that. It sounded from the post like OP was leaning towards Berkeley because of 1 or 2 big name professors and I just wanted to express my thought that going to Berkeley because of MJ alone might not be a prudent idea. Both are great schools and OP can't go wrong. Thank you for clarifying and offering your very helpful knowledge and perspective!

I just checked their webpages, and Jordan currently has 33 students and post-docs which is more than the entire stat ML group at CMU with like a dozen good professors has.

Agreed that these departments are both good enough that you shouldn't choose Berkeley just because of ranking - CMU has some amazing top people. I guess Jordan is in sort of another league for ML in terms of how widely cited he is, but I imagine there is quite a bit of competition to work with him and he has lots of students/post-docs, so I'm not sure making a decision based on one professor is a good idea. But CMU has people like Wasserman, Kass, Shalizi who are just as widely known as any prof at Berkeley. Agreed with above about cost of living. CMU has a generous stipend that will allow you to live on your own in Pittsburgh, but you will struggle to get by on the stipend with roommates in Berkeley.

Like half the department does probability stuff, so it's pretty hard to compare as a lot of statistics departments don't even have probability people anymore. Among their stats faculty, it looks like there's quite a few strong publications in places like Annals, so more theoretical work, but a lot of the faculty seem like they're on the older end and not as active any more. Some of the younger faculty went to top schools and worked with good professors, and they have some good publications. A lot of their webpages are broken or haven't been updated in years so it's kind of hard to tell. Seems roughly comparable to schools in the 50s or 60s on the US News rankings, except much more focused on theoretical stats.

I wouldn't think too much about the name brand as UW is perhaps the top biostat program. I'd also take cost of living into consideration as Seattle is incredibly expensive with tech companies driving up housing prices. UW has a pretty big department, so Brown might have a more personal feel to it. Hopefully someone can chime in with first-hand knowledge, but congrats on your choices and I'm sure you won't be able to go wrong.

I'm just a guy on the internet who has been reading the websites of some of these programs lately, so I don't have any first have knowledge of these programs. But I have worked as a data scientist at top companies and have some knowledge of what would make a good program. Some data science programs seem like decent options that allow you to take a mix of stats/CS courses (like Penn's program) and some seem like impressive career-oriented preparation (Northwestern's analytics program). Given the high cost of Duke's program and the syllabi of courses, to me it looks like a very expensive long bootcamp program (like Metis, Galvanize). To me, the classes look very superficial and you would probably come out of the program with decent Python knowledge, a very basic high level understanding of some statistics/ML concepts, and a project. I personally don't think that is worth over $100,000.

1. For statistics/biostatistics, you'll absolutely need to do well in a linear algebra class and preferably take a real analysis class. You should prioritize these classes over others. A probability class would be nice too. This is all you need and you'll be well-prepared. I'm not super familiar with applied math, but DiffEq would be more important there. 2. Generally (with many exceptions), biostatistics programs are a little more lenient with math requirements because their focus is on methodological and applied problems, rather than more theoretical stuff. Based on your post, I'm guessing you're more interested in solving applied problems in biology using math and statistics, so biostatistics would probably be a good fit. There is still PLENTY of math, but you won't be doing tons of theoretical proofs. Statistics and biostatistics PhDs are both great for jobs and are virtually interchangeable. Again, I'm not super familiar with applied math, but I don't think the job market is quite as good (because there are so many data science jobs now that like statistics PhDs). 3. A ton depends on your grades in linear algebra, real analysis, and if you can get a very high score on your GRE Q (167+ is ideal). The top programs (UW, Hopkins, Harvard) are probably out of reach since they can fill their spots with people with good grades from elite schools and research experience (which you should get if possible), and other Ivy League schools (Penn, Yale, Columbia, Brown) are pretty competitive, but there are many other good biostatistics schools to look at. I'd probably start by looking at schools from Emory down to UC Denver on the US News rankings, but you should post your profile here again after taking more classes and the GRE and people will be able to give you a more accurate estimate.

No harm in waiting til closer to April 15th to hear back from Michigan. I understand the urge to relieve some stress by settling on a decision right away, but in 12 days that won't matter, so if Michigan is a better fit (and a stronger program, probably) then I think you should probably distract yourself for a few days with something else, email Michigan in a week or so to check on your status and then make a fully informed decision.

UCLA is a solid department and it would be totally reasonable to go there over TAMU. Amini is good, and if you're interested in network analysis they have Handcock.

@remillion Congrats, awesome results. Can I ask if you retook the GRE Q or if the 156Q was the score you sent in?

$50k/year is basically the tuition at any US private school - Columbia is only significantly cheaper because it's 3 semesters instead of 4. Master's programs are where schools make a lot of their money. Essentially, they have to pay PhD students, and there are lots of sources of financial aid to make it affordable for undergraduates, but master's students are left out of most aid. If you plan on returning to Canada, I'd be pretty concerned about the ability to pay back over $100k USD in loans (plus living expenses). I doubt the school has much scholarship money but it never hurts to ask, especially if you have another offer like Rochester that might be more affordable? $150kUSD in loans would result in a monthly payment of like $1800USD/$2500CAD/month. Is there any possibility to attend a much more affordable Canadian program for your master's?

I think since most master's programs aren't extremely competitive, and you're coming from a good school with a decent gpa, you'll be fine. Since you're paying for the degree generally and not doing research, letters are much less important than for a PhD. Just some professors that you did fine in their class and maybe a future boss will be fine. There doesn't seem to be a whole lot of upside to getting the letter grades given that you're not sure you'll improve your GPA. I'd personally take the free opportunity to take all classes pass/fail without raising eyebrows.

1. Shouldn't make a difference. If you already have an account and can apply through that, it should be fine. 2. If they told you to submit it now, the previous deadline is irrelevant.

As you said clearly nobody will look down on you for pass/fail grades given the circumstances -- but if you are going to get straight A's (or at least A's in your stats classes) and improve your GPA, you should obviously do that. If you are going to get A-s or below, and thus not improve your grades, maybe you'd be happier relaxing?