bayessays

Brown doesn't have a statistics PhD program, so I'm not sure which program you're referring to (applied math probably? Biostat?) I definitely would not submit your subject GRE score. I'm going to be a lot less optimistic than the above. I think your 161 GRE Q is going to really hold you back. Yes, you have good grades, but grad-level course grades are known to be heavily inflated. As said above, this would be mitigated if you were taking grad classes at Harvard, but not so much if you're at Villanova (49). I could maybe see a big school like TAMU taking a chance on you, but I think you're going to struggle getting into a top 20 program with a 161. I would do whatever possible to raise that score - even a 163 would help you significantly. I think Washington, Cornell, Brown and NYU are not realistic targets. If you improve your score to a 165, I think they could be in your "reach" list and schools like Florida and PSU could be targets.

I'm sure you'd get into a top 10 program if you applied to all of them. I'd also probably apply to a couple of the bigger state schools ranked 10-20 like NCSU, PSU as relatively safe options. You won't have to go lower than that. I wouldn't worry about the third letter. If you have two strong ones from people who know you well, a letter just saying you are good at math won't hurt you. If you're dead set on going to Stanford, you'll need to take the math GRE, but even schools like Chicago don't really require it, so I don't think there is any reason for you to delay applying unless you would like to take a year off for fun.

Definitely agree with above, those schools you added and schools like Minnesota/Wisconsin should be the bulk of your applications, with a couple reaches like Duke. Penn State also has a great statistics department, so talk to professors there. Those last three schools I would say are in the safety range (though nothing is guaranteed), so I don't think you need to apply to many schools at that level.

What sort of state school do you go to? Berkeley vs Montana makes a difference. Also, it's hard to say without your GRE score and those future math grades if they'll be submitted. Do you expect to get a 165+ GRE Q and do you think you'll get an A in analysis? So much depends on these 3 questions that I'll refrain from giving too specific advice right now, but I'd say those top 3 schools are going to be your reaches (but perhaps obtainable) and the bottom 3 are probably low matches - there's a big gap where you skipped like 30 spots in rankings that are probably your best fit.

Measure theory will be more essential, but any type of analysis will help you if you want to do theoretical research. The exact subjects you will need to learn will depend on your research area. As an international applicant, I would recommend taking as much math as possible if you can get As. That will help you more than a non-research internship.

Your grades in calc and linear aren't great, but your grades in prob/math stat are even more concerning. Even if you got an A in real analysis, I would say Oregon State is the only school that's close to the realm of realistic and I would be extremely surprised if you got in there. You will probably need an MS to get into any reputable statistics PhD, but even then, it will be around the level of Oregon State. Those other schools are simply not realistic for someone with your math grades. Edit: to clarify so that this isn't misinterpreted as B+s being disqualifying, I am talking about the pattern of Bs in core, low level prerequisites with no higher level As to prove that you can do math. Lower grades early on are not disqualifying and common, but if your math education stops there, not good. OP would probably have to take a few semesters of real analysis/numerical analysis and proof based courses and get As in them all, and take a grad stat sequence and knock it out of the park to convince PhD programs he can handle the coursework

Almost all programs allow you to master out. Since you think you might want a PhD, I would definitely apply to PhD programs - your profile is so good that a master's will not help you. With your math background, high GPA at a school known for grade deflation, and amazing GRE score, I don't think you'll have to go outside the top 10 programs, with some safeties in the top 15. If you can get a very high score on the math subject GRE, that will help for a few of the top schools - if you can get a 90%tile, it would be worth it, since you definitely have the profile for the schools that want it.

I believe Berkeley's is in the school of public health, but heavily associated with the statistics department. The only well-known department I know of that is not in a SPH is Penn, which is located in their school of medicine.

If you can get a 165+ on the GREQ, as well as hopefully acing real analysis, you should aim much higher. Your profile is too good for schools like Drexel and GW, and schools like BU and Pitt would probably be your worst-case fallbacks. I'd apply for all the top 10 programs if I were you.

I think you have the right idea. I had an incredibly similar profile out of undergrad, but with some significant research experience that helped me out. If you're interested in biostatistics, I think they'll be more forgiving of your early math grades and you could apply to programs outside the top few with some chance there. Getting an A in analysis will help. If you go to a respected (preferably funded) MS program, ace it, get some letters from known people, you can really increase your chance at higher programs.

Also, a lot of internships only want people who are farther along in their studies so you don't even have the opportunity to do them every summer. I agree with above that 1 internship is common, 0 is also common. But I know people who have done 2 or 3 as well and if your advisor is supportive, it can be done.

I think you are selling yourself short, especially if you currently go to a decent (let's say top 60 US news) school. I'd throw apps at schools like Duke, CMU, and Michigan to make sure you have no regrets.

The Casella Berger course is the core of what a statistician does (and what will be on your quals if your program has them), and most people don't use much measure theory at all, so definitely don't skip that. It's hard to give advice without knowing the specifics of your program, but there is probably no way to speed up the coursework. If you want to finish quickly, the best thing to do would be to be an RA and establish a good relationship with your advisor so you can get started as soon as possible on your dissertation. But you're not going to be able to do any statistics research without taking a Casella Berger-like class, so just do well in your courses. Failing your quals by not focusing on coursework is one way to guarantee you won't graduate in four years. As for staying happy, I don't think putting an artificial timeline on yourself is going to help. This is going to be a slog, and it'll get done when it gets done. You can obviously try to go quickly (and this depends on your program - Duke is known for getting people out in 4 years) but I wouldn't want to be putting a strict timeline on this. My one tip is to try to have some friends and activities outside of your program. It'll drive you crazy if you just think about statistics 24/7 - if you're able to keep some perspective on the grand scheme of life it'll help take some pressure off.

I think that's pretty reasonable and seems to line up with what StatPhdNowPostdoc says.

Absolutely not. It doesn't make any sense to take the grad level real analysis without taking the undergrad course first, and it's totally normal to take it during the master's like you're doing. No program will expect grad-level real analysis. You're good!

Yes, I didn't mean to sound so heavy-handed. It's hard to give detailed advice on every single choice when the list is not in the order of difficulty. I don't think OP will get into Chicago or Columbia, but I could see them getting into Duke or Michigan or even JHU. I just think they should mainly be targeting those other schools and consider schools like Duke to be on the reach end.

Working backwards, to answer 3 and 4 quickly, I don't think you should retake the GRE and I don't think it's worth taking the math subject test - I don't see you getting into the schools that require it. On question 2, I think the first few schools are unlikely (especially Chicago) and that the Madison-UIUC part of your list should be your targets. On 1, if you could get into a top 5 biostat program, there will be opportunities to do theoretical research. However, your classmates will likely be very interested in applied research and you will be in a public health school surrounded by that. If you really just want to think about math, I think it makes sense to go to a statistics program if you want classmates that share your passion.

It would be very rare for undergrads to publish in journals like that, so it would definitely be impressive and help.

Yes, you have plenty of math. I think the ones you listed are a good start. A course on optimization may be useful as well.

1. Math courses will generally help more than applied stat courses. 2. Definitely take the advanced linear algebra. In fact, prioritize it over the two courses above. Even if you did well in your first linear algebra course, things like SVD are used all the time in statistics. 3. You don't absolutely need it, but if you will do well, take it. My ranking of priority for those 4 courses would be 1. Advanced linear algebra 2. Measure theory (Large gap) 3. Functional analysis/applied regression

I agree with you, and I think this person probably would have a shot at schools like that, which are ranked much lower than the ones listed. That's actually the exact range of schools Lp_space cited. I am by no means an expert on this, but I'd take the rankings with a grain of salt. Included in those rankings 701-750 is Smith College, the best LAC in the country - the top student at Smith is going to have a lot more success than the student from Universiti Utari Malaysia. The point is that at the top 15ish schools, there are plenty of students who have been vetted by Ivy League-level coursework, and it's a huge risk to take someone from a school in Canada you've never heard of. I know people who have graduate degrees in math from unknown schools with 4.0s who struggled in first semester PhD courses, so my comment was more about the fact that his school is unknown, not that it is in Canada. I suspect candidates from UBC, Toronto, McGill, Cambridge, LSE etc are not treated much differently than Americans, except for some funding opportunities.

Coming from an unknown international university, your profile is going to be scrutinized extra hard because they won't know the rigor of your coursework - for instance, they might wonder if that B- in measure theory is more indicative of your ability than the other coursework. You have a pretty good profile though, but your list is all extreme reaches. I wouldn't be shocked if you got into UCLA, but it's by no means a guarantee, and I don't think you have much of a shot at the PhD other schools. I think getting a master's at a well known Canadian school makes sense for you.

The difference between having this profile from an Ivy and having it from an unknown school is the difference between night and day.

Some schools ask for the textbooks you use, so taking a harder class and doing well will help you.

Your B- in analysis and your GRE verbal score are going to kill your application. Competition for international students is intense and I don't really see a clear path to you getting a PhD. If you are sure you want to get a PhD, my only thought would be to try to get into a low ranked master's at a somewhat reputable school, retake analysis, get a near 4.0, and improve your English so you can do better on the GRE but it's going to be an uphill battle.