bayessays

Yes, you almost certainly want to find biostatistics programs (there might be one or two faculty at other schools doing something with survival analysis, but you'll have to just dig through faculty pages and look for the words "survival analysis" or "reliability"). All the top biostats programs have people doing survival analysis - UNC, Harvard, Washington, Berkeley, Hopkins all certainly have big names in the field.

You say top 20-30, but your list is basically the 10 hardest programs to get into (except UC Davis, which I think is a reasonable target). Your grades are obviously good, but I don't think there's a lot that makes your profile stand out. Your math background and research are below average for people accepted to these schools. I'd classify all of these schools as big reaches, except maybe UCD which I would put in the "possible but not guaranteed" category. If you don't want to go into a research-track job, then going to a program in the top 20 is not by any means required. I think your best options are to either lower your expectations and apply to a much broader range of programs (there are many great programs all throughout the top 50 and even below the top 50!), or if you are not in a rush, get a master's and get some more background and research experience.

No, Penn State's placements are great. It seems like you for are putting too much stock in the rankings. Is there a single reason you can think of that would make you want to go to UNC? Because "jobs" or "it's a better program" are not valid reasons, because they're not true. Weather, location, smaller program, unique research interests (UNC has lots of probability people, someone doing object-oriented data analysis, and fiducial inference, which are unique research areas), wanting to live in the Research Triangle forever - these are clear reasons to go to UNC. Go to each school's website, go to every single faculty member's website, and look to see what they're working on and make a list of how many profs at each school sound interesting to you.

I think your assessment seems accurate - there is not a big enough difference between these departments where there is a clear choice to make based on their ranking alone. Especially if you're interested in doing applied work and working in industry, I don't think you'd see a difference in your outcomes at the two programs. For doing computational statistics, I know PSU has some good people like Murali Haran. They also have some cool applied areas (in the past they had astrostatistics, personalized medicine groups, etc). $11k+ a year is a lot of money - I'd look at apartments in the area too and see about cost of living. I think UNC could make sense if you want to set down roots, since you could theoretically stay in the area with the industry jobs in the triangle. The weather is also a little warmer.

UChicago's MS statistics FAQ currently says they give 10% scholarships to most people - even 50% still leaves a huge cost for attending the program, which is why the person probably sounded disappointed. Most people on this forum are applying to PhD programs and will get a master's degree for free, so any amount of money to pay would be disappointing when there are programs where you can go for free.

I can't comment on Applied Math/OR, but for statistics, I think your list is completely reasonable - if anything, I think Emory is shooting too low for you but NCSU/Northwestern/Rice are definitely the types of schools you should be end up attending at a minimum. It really depends on what you want to do though. Applied math and stats are quite different. You have a great profile for statistics at least though, and would have a lot of success applying to top programs if you applied to more stats programs. I doubt you'll get into Stanford, but I think you'd be at least in the conversation almost anywhere else.

I think you're shooting way too high with the PhD programs with TAMU and Ohio State (I can't speak to GA Tech's admissions as it's not a statistics program, though, and I have no idea how competitive master's programs are). They're ranked #13 and #24 in the US News. I think you should be looking more towards the 50s and below. You'd probably benefit a lot from a master's in terms of admissions. Your math background is fine, but the grades aren't ideal, and since you're coming from an average school with no research, I can't really see a top 50 program accepting you.

There isn't a single PhD program in statistics that requires measure theory, as far as I know. A few top ones will require measure-theoretic probability, which you got an A in. I can't image any scenario in which somebody looks at your math grades and even thinks twice about whether you're going to be able to complete the coursework and exams.

Not sure if you're trolling about your GPA. Maybe you won't get into Stanford, but you'll get into plenty of top 20 programs and probably quite a few top 10.

Both are good departments, so tuition is probably the biggest thing, so I'd really figure out if you can get an assistantship if that's going to make a big difference between the two. UNC hands down wins for the job market with Raleigh-Durham nearby. Madison is a pretty small city - they have 10 data science jobs posted on linkedin right now in the whole city. As for curriculum, 10 years ago I probably would've said that UNC's operations research/optimization focus is a little weird, but honestly I've seen so many data science roles lately that are looking for that type of background for their supply chains, so you might have an opportunity to set yourself apart there. If you don't have personal reasons to be in the midwest, I don't see many huge advantages to UW besides maybe the tuition and the fact that it's a bigger/(more fun?) city (and only you can decide if that outweighs the other stuff). Realistically, they are both great departments, so I don't think your choice will have a huge impact on your career - I'd prioritize which you think you'll enjoy more and leave you with less debt.

Anything - obviously a more involved project is better, but you just want people that are able to speak on your ability to work independently. But obviously also just choose people that you think will write the most positive letters. I don't know anything about this at all and I don't know if this is something you are interested in at all, but some schools may have special programs available to Hispanic applicants or may go out of their way to recruit more diverse classes. I have some departments in mind that are more explicit about this. A lot of programs will consider you for their master's as a backup, so you don't necessarily have to apply separately. You could always apply to a wide range of programs and try to see how the results go while you're still working, and then narrow down more after seeing how the first rounds of applications go.

In that case, your math background is actually really solid for most programs. I'd figure out your letter writers - even if you weren't super close, are there professors who stand out in your mind as being supportive? Even if you don't have "real" research, is there a professor you did a project for that could stand out? If you can throw together 3 decent letters, I'd apply a little higher than what I stated above, and throw in some master's as a backup. I definitely think you belong in a program that's higher than what I stated above, but you just gotta put a couple pieces together in terms of selling your profile.

What type of school did you take the math classes at? Was it an online university? Your biggest issue obviously is that you don't have any letters of recommendation, so that will make it very difficult to get in anywhere since there isn't any research or pedigree that makes you a standout candidate. In your current shape, I'd probably start the search at US News rank 60 and below for statistics PhDs, and 25 and below for biostatistics. You would benefit a lot from a master's program where you could form some relationships with professors, get some research experience, and drastically improve your results.

Think about it this way: if you want to study probability because it's only really a research area at MSU, then who will hire you afterwards besides MSU? Most departments have no interest in hiring people who are not doing statistics research. Overall, I think Florida is the best department by a pretty wide margin. I don't think there's a huge difference between ISU and MSU, but they both have a few good professors who have students that get faculty jobs.

NYU does not a curriculum that will help you get into a PhD program at all, so if that is a concern, choose UNC.

Have you ever lived somewhere that's not California? Seattle's weather concerns are overblown - it's less rainy/cloudy than many major cities on the east coast, and the winters are relatively mild. It's not like going to Chicago or Minnesota. But, obviously it's not Santa Barbara, and if weather is a huge concern for you, then only you know how much that would affect your happiness. I think the best thing is to really think about your goals. You should be able to get a decent job coming from any stats PhD program, but Washington is really in a different tier than the other two and I think you'd have more options, especially if you're interested in machine learning. I think that holds even if you want to move back to California - you'l have more options in San Francisco after attending Washington due to the prestige of the program and your opportunity to do cutting-edge ML stuff. On the other hand, I'm looking at UCSB's PhD placements and they are still pretty good: https://www.pstat.ucsb.edu/graduate/alumni Location is huge though, and it's 5 or 6 years of your life, so if you think you'd be much happier in California, you can definitely be successful at any of the programs.

Building on stxnre's observation, I think that you basically have to think about the financial situation in terms of what you'd have to give up. I think you'd be able to have a nice studio apartment downtown and eat out regularly in Ann Arbor, versus Berkeley where you'd almost certainly have to have roommates and you might not have the money to enjoy the area as much.

At least in the near past, people with a Master's usually finished a year faster at Michigan. I wouldn't worry about anything you mention in this paragraph - if you're working on your dissertation at either of these schools, I can't imagine any situation in which they don't fund you unless you start taking like 8 years. Obviously the warm weather is a huge factor, but Ann Arbor is a very nice place to live. I'd visit both if you can. They're both good programs - I don't think you can go wrong. Both are schools that have good stat programs with lots of faculty to draw from - Berkeley does have stronger ties to stats department just because of the way the department is designed, but there are definitely people in biostat at Michigan working with stat faculty. I'd ask students about the requirements - a 20 hour a week teaching assistantship is often 5 hours or less, at every program I've ever seen. Once you start your dissertation, your "research assistantship" is usually just doing your dissertation.

All of those departments are prestigious enough where, even if you would want to go into academia, I think you could easily choose any of them. It sounds like you want to go to UT-Austin. Read what their recent PhD grads are doing, lots of good data scientist jobs: https://stat.utexas.edu/alumni-and-friends/phd-and-postdoc-alumni

I think you're right about Chicago's courses being more theoretical - from what I can tell on their website, their MS students have a lot of overlap with the PhD students' classes, and their PhD program is intense. I don't think this means you'll be less prepared for industry (you'll have to do a lot of stuff outside your classes for that, anyways - you absolutely cannot rely on a short Python course to learn coding for a data science job), but it would mean your life would be harder while in the program. I think all the programs you've listed are fantastic schools and I don't think you could go wrong with any of them. Chicago may have a little more of a "wow" factor, though.

Obviously I understand your desire for privacy, but can you explain a little more what a "solid" private school is - I'm guessing you don't go to Stanford/CalTech, but is your school on the level of USC or Pomona, or something like Gonzaga? I think your reaches are pretty far reaches given that you don't have much research and (I'm assuming) don't go to an elite school. From the limited details, I'd say: Reaches - NC State, Texas A&M, UT Austin, Matches - Illinois, Virginia Tech, Virginia, George Washington, Boston, George Mason, UMBC A lot probably depends on what type of school you go to and how substantial your research this summer is and the type of letter you can get from it. I know you're a math major, but you didn't include a lot of details about your exact coursework. If you took the minimum required math at a low-ranked school, that will be very different than someone who took a bunch of high-level or graduate-level classes at USC. You may want to look at biostat programs too.

Your other program is not forcing you to make a decision until April 15th presumably. I'd wait as long as possible to accept another offer, and I'd be expressing my strong interest in Stanford even after April 15th if it's your first choice, even if that means later reneging on your original decision. The only real negative to waiting is that you are left in a state of limbo for a while and you would have to wait to sign a lease for the fall.

I've never heard of an interview asking linear algebra questions anywhere, but anything is possible - you should really just be prepared to talk about your research and your interest in the program. I had an interview once that got a little technical about statistics stuff, but I was coming in with a master's degree and professional work as a statistician. It's really going to come down to the individual interviewer - there are probably weird people who are going to quiz you about irrelevant math, but there's not really anything you can do to prepare for that anyways.

I think your research is fine - research is a big commitment, so I don't think it's worth it to go out of your way to get publications or anything like that. If you have three letters from tenure-track professors in stats/math at one of the top schools in Canada, and they are very positive letters, you're in great shape. A master's can definitely help, especially as an international student, but I don't think it's absolutely necessary. If I were you, I'd just expand your list - apply to additional schools like UT-Austin, Ohio State, Illinois, that are great programs but not at the very top. See how your results are, and re-evaluate if you need to.

You've basically listed all of the most competitive programs in the country. You might have a chance at them, but this cannot be your entire list of schools. Your profile is strong in that you have good grades from a good school and a a very strong math background - your research experience is sufficient, but without publications doesn't really stand out (at least from the details you've provided). I'd widen your range through the top 30 on US News.