statsguy

I had the same dilemma a few years back and ultimately chose grad school. One point of clarification: you will mostly likely not be paying for the PhD in Statistics. You will receive a tuition waiver and 15-20k/yr stipend for being a teaching assistant. Also, when you are applying, the standard in the field is to apply for the degree you eventually want to get. So apply to PhD programs if that is your end goal. You have the option of picking an MS up along the way. Some schools award it automatically after passing the Oral Prelims. This is true for most PhD programs (except maybe those in humanities.) My reasons for choosing grad school: 1. The actuarial job market is very crowded right now, especially entry level. 2. I want to get all my education done with ASAP. Having a PhD at age 26 and never having to worry about school will be nice. 3. (The big one) More flexibility with a PhD (for example, you can work in Biostatistics with a PhD in Statistics, but can't with just a BS + actuarial exams.) 4. I really enjoy Statistics (both Theoretical and Applied) and want to really gain a deep understanding of the subject, rather than just know what's required for actuarial work. I know one person who became and actuary and is going to be applying for PhD programs this fall. Another person I know who became an actuary really enjoys the work, makes great money, and may one day go back for an MBA. Stats grad school admissions are becoming very competitive and if this trend continues (more applicants, fewer funded spots), you may have a more difficult time getting in later. Just another thing to think about.

Most application deadlines are mid December while some are mid January. So you will probably have to ask for recommendations in early to mid November. It doesn't hurt to apply for Fall 2011, but IMHO you will have a much easier time applying (and will likely get in to many more and better schools) if you hold off another year. Masters funding depends on the school; some schools fund a lot of their MS students for most of the time, some don't fund at all. Lots of departments are having budget issues, so funding may be harder to come by. My department (a very good department) will not fund any MS students for the next few years since we are struggling to fund PhD students. In the past, we have been able to fund at least 1-2 semesters of the MS degree.

It depends on the school. Some schools, such as Stanford, require the subject test, so I'm sure a high score would be helpful. Other schools recommend it, so a high score might help, especially if you come from an unknown undergrad university. It can also be used as a tie breaker. In my experience, it seems that most schools don't care about the score, or if you even took the exam (even some top 10 schools.) Although a good score definitely won't hurt, you would be much better off spending the few months doing a small research project in Statistics and expanding your knowledge in the field, rather than studying topics you will never see again (ie Complex Analysis, Topology, Logic, Abstract Algebra, Number Theory etc... ) for a really difficult exam. So I would suggest that unless one of the schools you are applying to requires it, spend time on research. It doesn't have to be a really deep, extensive, theoretical project; even some data analysis for a professor would look great on an application and would give you some insight into the field. It would also give you something to talk about in the admissions essays. And do not underestimate the difficulty of the Math GRE subject test. It's an incredibly difficult exam, considering the range of topics on the exam, as well as the time you are given to complete the exam. I've known some incredibly bright people would studied for 6+ months, and just barely broke the 70%ile.

Other points: 1. Apply to a LOT of schools this year, due to the budget and competition issues mentioned above (at least 8 schools: 2 reach schools, 4-5 mid schools, 2 safeties. AT LEAST) 2. Letters of recommendation are VERY important. Choose your writers carefully. 3. To emphasize once more, you are much better off going to a mid-tier school with research that interests you and an environment that you fit into, rather than a top-tier school with research that doesn't interest you/bad environment. 4. It will take 5-6 years to finish the degree, so you may want to consider going to school in a bigger city rather than a college town like Purdue, UIUC, Penn State etc.... Just somthing to consider. 5. Good luck!!!

5-6 years is more accurate. It depends on your background (for example, if you already have a Master's degree.) It's pretty rare to finish under 5 years.

DO NOT go by rankings! Forget about those. In the stats community, there is essentially no difference between the two programs in terms of respect/prestige etc... In fact, I'd say UChicago is more respected and highly spoken of. I know you probably made the decision by now and I really hope you chose UChicago. Do not let the name fool you when it comes to graduate school. Lots of big name schools have some departments which are weaker (e.g. Northwestern Statistics). That's not to say Harvard isn't a great department (it is), but it's so small and the research there is so narrow and concentrated. UChicago has excellent job placement and more diversified (although highly theoretical) research. Please please please forget about the Harvard name when looking at the program. Look at the programs blindly without regards to name. If you are looking for a faculty positiog afterwards, remember that after a certain point, the quality and quantity of your own research will trump just about everything, including where you went to school.

You are asking quite a lot in your post so I will try to do my best to answer some of your questions. If you want a job in econometrics, you can get it by getting an Econ Phd and focusing on Statistics, or by getting a Stats PhD and focusing on Econometrics. Both are probably equally competitive in the econometrics job market, but I think the Stats PhD will give you a deeper understanding of Statistics and will also be more flexible in the job market (if you ever want to do something else.) Choose your schools carefully.

A terminal masters will not get you very far in most Biostats jobs, especially in settings such as big pharma. You may be able to get a decent entry level position, but there is a very low ceiling when it comes to advancement if you don't have a PhD. A friend of mine got a job in pharma with an MS in Stats, and while his pay is pretty good (I think around ~60k), he will pretty much be stuck being a SAS programmer forever and is considering going back for his PhD or maybe an MBA. Most PhDs will start around 80-100k and have unlimited potential for advancement. For finance/insurance jobs, having an MS is sufficient. I've heard pretty good cases for getting a PhD in Statistics vs. a PhD in Biostatistics. If you are absolutely sure you want to stay in the public health/biostats realm for your entire career, then a PhD in Biotatistics is probably slightly more advantageous than the PhD in Statistics. However, when it comes to other more industrial engineering/finance type jobs, a PhD in Statistics is probably more favorable. It's also pretty easy to get Biostats jobs with a PhD in Statistics. Several former students in our department have gone on to work in big pharma/public health with a PhD in Statistics. So to answer your question: yes, the PhD in Statistics is generally a little more flexible. Edit: As a warning, the curriculum for the PhD in Statistics is usually much more difficult and goes into a much deeper level of Mathematics than the Biostats PhD cirriculum, so keep that in mind if you don't have a strong background in Math (ie Real Analysis, Advanced Linear Algebra, Measure Theory etc...)

I'd go with a PhD in Biostatistics. It's much easier to get a job in any branch of Biostats with a PhD, and the pay is much higher. Also, if you want to advance up the ranks, a PhD is absolutely necessary.

I mean, what can anyone really say? Unfortunately, I don't think there is a whole lot you can do except hope that someone declines their offer. Do you have any other offers that are still valid? I guess contacting the grad program director to plead you case can't hurt, but again, since it's been almost two weeks since you "declined" your offer, I don't know they'd be able to do. I would immediately contact all other schools you declined to see if there is any possibility of reactivating an offer. If you just need to go somewhere, you may have a chance. Otherwise, if you are set on Stanford, you may have to wait until next year.

This is completely untrue when it comes to fields such as Computer Science, Mathematics/Statistics, Physics, Biology etc... Most students are accepted directly into the PhD program in their field. A master's degree can be picked up along the way, if desired. Some schools give it out automatically when the student passes their preliminary oral exam. Also, to the OP, Real Analysis freshman year? At a top university? That is very impressive. Out of curiosity, why are you taking Organic Chemistry if you want to get a PhD in CS? Seems like you'd be better off using the time to get started on research or take additional CS classes. Or enjoying life, partying, boozing, smoking weed, or (insert fun activity here). Two of my roommates took Organic Chemistry in undegrad and it was a total time suck. Focus on acing your CS classes. Get started on research ASAP. A close fried of mine is getting his PhD in CS at my school and said that, by far, research will help you the most when it comes to admissions. LORs are a close second. Grades and GRE Subject Test is third (this was similar for my field and is for most sciences.) Don't worry about non CS courses. You just have to pass those. Grad school apps will want a subject only GPA (for CS classes), and an overall GPA, which they could care less about unless it is extremely low. Best of luck to you.

I'm kind of torn on the issue. On one hand, it would be nice if someone told you exactly why you weren't admitted. But on the other hand, what would the department tell you that you don't know? There are really only a few factors that ADCOMs consider: GPA, LORs, GREs, and research. Obviously one of those was lacking (although I highly doubt it was your Math subject test score). If you will not be reapplying to one of your schools, then I guess it can't hurt. Just don't expect a detailed response. Keep in mind that in the past few years, applications to grad school have substantially increased. For example, my department had 80 application for the PhD program in 2004, but received over 200 this year. On top of that, we were only able to take 5-6 students rather than the usual 8-10 due to budget issues. This is happening in most schools - universities are not immune to economic downturns. It could be that had you applied a few years back, you would have easily gotten in, but due to increased competition and fewer funded spots, you may have just missed the cutoff. Lots of rejection letters this year have mentioned this. Also, why only apply to four schools (all of which btw are extremely strong programs in their respective fields)? I'd recommend if/when reapplying, send out at least 8 applications, with two of those being "safety" schools. Good luck to you.

When I applied to PhD programs a few years back I got one of my LOR writers a case of his favorite imported beer, another a premium, top shelf bottle of imported Cognac, and the third I got a gift card to Starbucks. All of them went wayyy out of their way to help me get research money in undergrad and fellowships for grad school, so I figured they deserved a little extra something. They were all delighted. I'd say be personal. Two of my Math professors that wrote me LORs were beer judges and pretty heavy boozers. The third loved coffee so put it all together and you have a solution.

That's the one thing about Stats departments: there are the 5-6, top-tier schools (Stanford, Harvard, UChicago, UWashington, Berkeley etc...) and then there is a huge cluster of very good programs (in these rankings, they would be the schools rated approx. between a 3.5 and 4.0) It's really hard to differentiate the schools in that area. What was suprised me was NC State's ranking, which, after removing the preceeding Biostats rankings, comes in around #7. I'm really glad that USNews seperated Stats and Biostats, although it would have been nice had they created two seperate lists. By the way, congratulations on OSU. I actually recieved the first year fellowship their when I applied, but did not end up going there. A really solid department.

Reputation-wise, I'd say they're pretty much the same. Both have some excellent faculty, as well as great job placement. Also, both are in pretty good locations. If it were me, it'd come down to research.

Hands down Penn State. UCLA, despite their name, does not have a strong Statistics PhD program (their Biostatistics program, on the other hand, is excellent.) Additionally, I have heard quite a bit about how poorly it is managed/run. The graduate students I spoke to that go there seem to be fairly frustrated and unhappy. And judging by how they handle admissions (which is when programs are usually on their A game), I don't have a reason to believe these rumors are untrue. Also keep in mind that the cost of living at UCLA is much higher than in University Park, but I think UCLA's stipend is only slightly higher (maybe $1k more, but I don't remember exactly.)