JZappa

If you are talking about UMD's Mathematical Statistics program I would definitely recommend UVA. UMD's program is very theoretically oriented from what I've seen and is much more on the math side of things than on the application side. I could be wrong but that's the general impression I've gotten from what I've seen.

Out of curiosity what job statistics are you using? There was a study by the American Statistical Association that shows salaries are quite a bit higher. Also why don't you ask if your PhD program gives an embedded masters. If it does you could just drop out after you get the masters and you'll have been funded the whole time if you decide that's what you want to do.

I'd say the biggest thing is that the lack of analysis for the very top PhD programs is going to be a big turn off. I would say that an MA from Berkeley should help considerably though. Just make sure that you talk with professors to get a good idea of which classes to take to make you more competitive for PhD programs.

I disagree that UNC is just Probability, OR, or ML. I'm a student there now and we have several really good professors working on all sorts of applied things. I know we at least have a couple people doing applied statistics (Prof. Ji comes to mind), time-series, fiducial inference, and a lot of people doing biostats, and some other related fields. Also UNC is definitely not rural. Raleigh is a 30 minute drive away, there is RTP (which you can use for funding opportunities i.e. lots of available internships), and Durham is maybe a 5 minute drive as well. Plus the fact that we have SAMSI means there are an abundance of chances to do good applied work with a lot of other people. If you can definitely visit campus and talk to people about things, applied stats isn't really my area so I can't completely attest to exactly how good we are but it seems like the majority of our non-probability/OR professors are very applied.

Northwestern is a really small department and they already had some admits in late Jan/early Feb. Chances are they don't do rejections until all their spots are filled so you're probably just on a wait-list (where on the wait-list I have no idea)

Usually I think the answer is yes. You might have to have an advisor in the biostats department but I know it is at least somewhat typical to have a coadvisor in another department.

My bad I forgot about Professor Marron, I don't see him around too often. I was only counting the stats department and I am pretty certain those are the only 3 major machine learning people, although there might be a few others that dabble in it.

You definitely need linear algebra and a real analysis course (sometimes called adv. calc) to be on the safe side. Also a course in probability would be really helpful as well. If you can get those classes in and do well in them I don't see why you wouldn't have a good shot at getting in... Stats is very interdisciplinary so I think a fair number of stats students don't have a math background. I know one PhD student at my school who got his undergrad in some kind of political science and then worked for Oracle for a few years before enrolling in the program here so it is most definitely possible.

Luke8ball... I wouldn't email and specifically ask if you are on the waitlist but it is perfectly acceptable to ask what your status is at this point. Just be polite and ask them if they can update you on your status or when you should expect to hear something from them and you should be just fine.

So while I've been accepted to several stats departments I believe my two best choices are UNC and Michigan. The problem I'm having is that I am currently an undergrad in the UNC stats department and so for all the typical reasons would like to go somewhere else for graduate school. That being said I'm very interested in probability and some of the more math oriented theory (stochastic calculus for example) and UNC has several fantastic professors for this sort of thing (Pipiras, Budhiraja, and Bhamidi) all of which I know very well and have done research with. Michigan doesn't quite seem to compare in terms of their probability group, most of their faculty on the web page anyway seem to be older and close to retirement or not as active (keep in mind I'm solely talking about probability). On the other hand I'm also interested in machine/statistical learning which Michigan does extremely well and which UNC doesn't do so much of (only 2 professors). Probability is still my primary interest but Michigan seems to have a better overall department in terms of breadth and depth of coverage into other areas. All that being said if anyone has an opinion on what to do in terms of picking between both of these schools I would greatly appreciate it.

biostat_prof... I thought Yufeng Liu was in the regular stats department at UNC. Or does he have a joint appointment?

Most schools (UNC Stats does this) don't send out rejections until they have filled all their spots. In a sense you're on a very long waitlist.

I wouldn't be too concerned just yet. If you haven't heard anything then you at the very least haven't been rejected yet which seems to imply you're either still under review or on a wait-list. Either way I think it's not that uncommon to get off a wait-list.

Econ_to_Stats. I wouldn't worry too much I know for UNC-CH at least they've barely started doing things on the PhD end, only a few admits yesterday and from the professors I've talked with they've been moving pretty slowly this year (which I get the impression they do every year). Given that PhD students tend to hear things earlier than masters then I would guess they'd send something out in the next month or so? Spring break is coming up otherwise I'd say it would be sooner. Unless Bowling Green didn't offer you funding is there any reason why you would go to a masters program over a PhD program?

Yep never fear! I had a semester that was quite poor and pulled down my GPA but at least thanks to the statement of purpose or what not I was able to explain it and I've gotten offers from several very good stats phd programs. Not that that is the same as math but still departments definitely take the whole picture into account rather than just your numbers. Also if you've done well in all your math classes which it appears you have then they aren't going to care about a few lower grades on the business end of things.