cyberwulf

When I say someone is "competitive for PhD admission", I mean they are a strong candidate who will be given serious consideration. Of course, not all strong, seriously considered candidates are ultimately offered admission but many are. I stand by my evaluation of the OP's chances; while their overall GPA is 3.8, their math/stat GPA is ~3.9, which is a very strong record coming from a major state university like UW.

Yeah, the verbal GRE score really isn't a big deal. If you're a U.S. citizen, you'll be competitive for PhD admission at every program in the country. If not, Harvard/Hopkins could be tough to crack but I'm guessing you're still pretty much a lock at UW and are in a very good position for everywhere else.

You will be competitive for admission at all programs. I would just take the on-campus LA course in the fall, and send an updated transcript in December; most schools don't get around to Masters admissions until around February anyway, and I wouldn't be surprised if some places took you before seeing the grade.

Not any decent ones that I know of.

I think this might be better handled in the domain-specific forums, since whether or not these contact letters are required/expected varies greatly by discipline. In my field (statistics/biostatistics), they are completely unnecessary and a waste of time.

As an addendum, the one main exception to the "personal statements aren't important" rule is applicants who have an unusual academic record/background compared to the typical stat/biostat grad applicant. This could include, for example: - Applicants who have been out of school for several (5+) years. - Applicants who are changing fields. - Applicants who lack prerequisites. - Applicants whose academic performance was affected by serious personal/medical circumstances (e.g., one semester of terrible grades due to death of parent, major illness, etc.) If you fall into one of the first three categories, we want to know why you chose statistics/biostatistics and how you think your background has prepared you for success in grad school. If the last category applies, it's important for us to know since it provides needed context for interpreting your previous academic performance.

Every admissions season, many students applying to statistics and biostatistics programs are intimidated by the task of writing personal or research statements. Indeed, there is an entire sub-forum on GC dedicated to SoPs (Statements of Purpose) where there is much hand-wringing over how to craft the perfect text. But while I can't speak for disciplines outside of the statistical sciences, I can confidently say that in stat and biostat, the evidence strongly suggests that personal statements have little impact on admissions. I've written several posts about this in the past; here's a summary of why you should stop worrying so much about a 1-2 page essay: 1. Mathematical ability is best assessed through academic records and test scores (and to a lesser extent, letters), so it is generally quite easy to order students on this important trait.The pool of students applying to statistics and biostatistics departments isn't particularly deep, so that a major concern of even excellent departments is whether applicants can handle the requisite mathematical coursework and exams. 2. Very, very few applicants have meaningful statistical research experience before starting graduate school. As a result, many students end up working on dissertations in areas entirely different than they were initially interested in... and this is totally OK! 3. Funding in most (but not all) U.S. stat/biostat programs is allocated at the department level to the strongest incoming students, so applicants aren't typically "matched" to potential advisors who agree to fund them*. Rather, the department projects the total number of positions available and then tries to recruit up to that number of students. Once the students are on campus, they are then either assigned to a position or (ideally) have some choices available to them. Given points 2 and 3, declarations in the personal statement such as "I am very interested in studying [X] with Professors [u,V,W]" usually carry little weight. They typically translate to: "[X] is a hot topic which I know very little about but sounds interesting, and I see on your website that Professors [u,V,W] list [X] as a research area." Which, again, is JUST FINE, since that's essentially all most people can credibly write. 4. Research potential *is* important, but the best source of information on this trait is letters of recommendation, not a one-page essay. In some fields, part of showing research potential is demonstrating that you have already thought of a reasonable project that will turn into a dissertation. Since (virtually) no one applying to stat/biostat has a "shovel-ready" dissertation idea, research potential is generally assessed using some combination of mathematical ability, creativity, and perhaps some exposure to lower-level research, all of which are best evaluated using other parts of the application. I don't mean to denigrate the personal statement too much. There are a few key things to avoid (eg. rampant grammatical errors, aimless rambling, saying you have no intention of pursuing an academic career if you are applying to a PhD program) and of course there will be exceptions to every rule, but in general, as long as the PS is competent it probably won't affect your chances of admission significantly.

Possibly C, but only for academic researchers. If you're doing stat genetics or big data, languages like Python and Perl could be handy to pick up.

Totally agree with this advice.

At MINIMUM, you will need the following math classes to even be considered: Calculus I, II, and III (up to multivariable) Linear Algebra A year of probability/mathematical statistics You will also need Real Analysis to be taken seriously by anywhere halfway decent. Even with those courses, you'd have pretty marginal preparation. Most applicants to stats programs were/are math majors, and so will have taken quite a bit more than the above. The graduate statistics courses you could take with your current background are likely very applied and are not going to help you much when applying to stats departments. So it's a bit of a tough road, but the good news is that if you have 3-4 years, you should be able to pick up the relevant pre-requisites through a combination of in-person and online classes.

Not necessarily. The bootstrap can do a pretty bad job of estimating distributions of extreme order statistics (i.e., min and max). I can't think of an obvious way to do this; perhaps the reason that Van Geert's technique rarely yields a small p-value is because it (correctly) accounts for the unstable behavior of mins and maxes.

If you continue doing well in your classes (especially your math classes), you will be a lock for any biostat Masters program. And, if you keep up a GPA near 4.0, you will be pretty competitive for many top biostat PhD programs. You might consider checking out the SIBS program next summer to get some more biostat exposure -- there's one at Wisconsin, and a couple of others in the Midwest if you're not looking to venture too far afield.

I think that a Masters in CS might help you, particularly if you end up looking to land a job at a place like Google/Facebook.

Practically anything is better than the math job market... The job market in stats is a lot better than math, certainly. While graduating from UCB doesn't guarantee a faculty position at a good place, the outlook is still pretty decent since the private sector provides a fallback option.

My general advice is to go with the best/most prestigious program that will accept you, in a place that you would be happy living. There's been a lot of discussion on this board about what people with particular academic records can aspire to in terms of admissions, so I suggest you try to find a close-ish match to your profile and start to form a list of reach, reasonable, and safety schools.

Uh, surely you're trolling? You have a 3.94 in math from an ivy league school and you think you're "screwed" if you apply this fall? Unless your letters are horrific, you have a good chance of getting into any top PhD program in statistics.

Interesting case. The answer depends on what you mean by "Tier 2" or "Tier 3" biostat program. If you're getting an MS at Michigan or Minnesota, then a ~3.9 GPA will open a lot of doors, and should put you in decent shape for admission to programs at their level or below. If you're at, say, Ohio State or Iowa, then it might get a bit dicier (though certainly not out of the question) for places in the top 10.

Purely based on a quick scan of the list of faculty research interests on the Baylor stat website, I don't see a lot of people who claim to primarily focus on biostats. How did you come to this conclusion?

In theory, "expanding your knowledge base" is a nice idea, but in practice these labels don't mean much. People hiring for academic and industry research positions care about what you've done, not the words which are printed on your degree or resume (this is because, unlike entry-level hiring, there are usually a small enough number of qualified applicants that each one can be evaluated in some detail). To give an example, someone graduating with a PhD in "vanilla" comp sci working with an advisor who is well known in statistical machine learning would likely be better positioned to land a stats-oriented position than someone who completed a stats co-major but did their dissertation in, say, operating systems. I'm all in favor of learning about areas which will make you a more well-rounded researcher, but do it because it gives you essential skills, not for the sake of being able to check off a box on your graduation paperwork.

I would counsel against committing yourself to this. Things like co-majors and graduate minors carry virtually no weight when you're on the job market after graduating, and they limit your course options while in school. There's presumably nothing stopping you from taking those additional courses even if you're not getting an official "co-major", and if you take a couple of stat courses and decide it's not for you, you won't have to slog through another one just to complete the co-major requirements.

Nobody pays attention to what college/school you graduated from within a particular university. What matters is the courses you took, and how you did.

Well, if you like hiking and the outdoors, Seattle is pretty tough to beat. Though there's no doubt they've got a bit of a sunlight problem. While quality of life *is* important for a graduate student, I generally advise people to keep an open mind geographically and not let preconceived notions weigh too heavily on the decision-making process. For most, the key to a happy grad school experience is being in a department where they feel motivated, challenged, and encouraged. Though I have no evidence to support this, I suspect that students who make grad school decisions based mostly on location and other factors unrelated to department quality are the most likely to be disappointed in their choice and eventually burn out.

I think your profile will play well in biostat and comp bio departments; you're probably a little thin on the math side to crack the top stat departments. I don't know much about comp bio/bioinformatics programs, so will focus on the biostat side, where I think you will be competitive for admission at all the places on your list. Strongly encourage you to consider applying to UW, since they seem to be going "all in" on the big data wave, and have a lot of exciting young faculty. And if you're interested in Bayesian methods, most of Minnesota's highest-profile biostat faculty are Bayesians. My biostat list for you would probably be: UW, Hopkins, Michigan, UNC, Minnesota, + 1-2 safeties (I think UCLA and Emory qualify, given your profile) For stat, I would throw apps at NC State and CMU. I wouldn't bother with Colorado State.

If Theory of Computing is about closures, lambda calculus, and other theoretical aspects of programming languages, then it won't help you for an MS in stats. If it's about running time analysis of algorithms, big-Oh and little-Oh notation, etc. then it might be marginally useful. I took PDE as an undergrad, and have never used it since in my statistical career, but I think it would be pretty fundamental if you're thinking of going in the direction of applied math.

I think that Emory is the better program. Its name is well recognized within the biostats/public health community.