Statistics PhD Profile Evaluation

[huxlb]

@bayessays What would be an example of a school with better funding? I wasn't aware of the funding issues at Georgia. I'm mainly looking at schools on the East Coast and Midwest for geographical reasons. My research interests are pretty broad right now, but I like UConn because they have some professors working in Bayesian and computational statistics, which interests me. I'm not sure if I want to focus on solely biological applications, so I don't know if biostatistics would be the best fit. 

[StatHopeful]

Just to add something to the decision:

I think @bayessays is right, that you should replace some of the lower programs on your list with some of those mid-ranged ones. 

That said, I'm not sure UGA has funding issues. I just returned from a visit day, and I know there were 5 students domestically who had been funded and were able to attend while there an additional 7 internationals who had received funding. IIRC they have had a few students begin the program without funding and picked it up after the first year. It seems that hardly any of these funded students are on the results page, however. 

If you really like UGA's research or want it for some other reason (location, known a professor, etc) then it wouldn't be horrible to apply. I think you'd probably get funding there.

Edited February 19, 2018 by StatHopeful

[bayessays]

If you look at Georgia's results on the results page, they admit a bunch if PhD students without funding which is almost unheard of - but you should decide if the positives of UGA outweigh those. Sounds like it wouldn't be an issue for you personally, but it doesn't exactly reflect well on the department. If you really want to live in Athens, then go for it.  Besides Duke, Mizzou, VaTech, Iowa, UT-Austin are very Bayesian. OSU has some Bayesians and is a large strong department. On the west coast, UCSC is extremely Bayesian and UCI is pretty strong there too.

[huxlb]

@StatHopeful @bayessays Thanks for the input! I was just considering UGA as a safety school, but will definitely re-evaluate that now. 

[StatsG0d]

I think Duke and CMU are far reaches. I think you'd need to improve your GRE scores drastically to have a fair chance at either of those programs.

The other schools seem to be ok. I'd definitely ex off UCONN. I applied there in 2015 and they tried to admit me after April 15. I haven't heard if things had improved or not, but I think it's a risk and probably not worth the money investment.

My biggest concern with your profile, however, would be the writing score. A 3.5 is not very good for a domestic applicant. And while math matters more than writing, there is at least one program (Penn State) that looks at the writing score heavily. Also, it might send a message that you cannot write analytically, which would not be a good sign for papers. This is my opinion, however, and mostly speculation.

If i were you, I would consider retaking the GRE and aim for slightly improving your writing score (even a 4.0 is much better than a 3.5). When I took the GRE the second time, I improved from a 4.5 to a 5.0 without even studying for the writing portion. So study for the writing a bit, study for the math even more. Get that GRE up to a 167+ and those two programs I mentioned above are much more attainable. That being said, I had a 164 on my GRE-Q and got into some top-5 biostatistics and some top-20 statistics programs, including several of the ones you mentioned on your list.

[huxlb]

@footballman2399 So if I were able to get my GRE up, are there any other suggestions you’d have to improve my profile?

[StatsG0d]

I definitely think you have a strong profile and I think even a marginal improvement in GRE could have a big benefit for you as it's the only true "weak" spot on your profile. Everything else looks very good to me. And as others have pointed out, the research experience makes you a unique candidate as very few people will have had that prior to graduate school.

If I were to suggest something, it would be to target maybe one more of the larger programs (e.g., PSU or Purdue). I noticed others were talking about Northwestern. NU is difficult to crack even though it's ranked relatively low. Several people (including me) were rejected from NU despite getting offers from more prestigious programs. It brings up the question whether it's worth it to apply, or to invest your money elsewhere. If it were me in hindsight, I would have saved the NU money and put it towards another reach school that probably has a similar acceptance rate anyway. 

Also, I would check what the research strengths are and what you're interested in. For example, UNC and NCSU are both great programs and in a great part of the country, but UNC focuses mostly on probability theory and is on the smaller side and NCSU is a heavily applied department and the largest department in the country. If you know for certain you won't be interested in only applied or only theory, it might be wise to cut out some of these schools. If you have no clue what to do, your best bet is to apply to programs that are large and have strengths in both applied and theoretical areas. A&M, Purdue, and Penn State come to mind. Florida is a fantastic program if you're interested in Bayesian MCMC methods, but significantly less attractive if you're interested in almost anything else. You also might want to consider some biostatistics programs. You have loads of time, so do some research and see what sounds interesting to you. If you decide that Bayesian isn't your thing, then you can take Duke and Florida off the list and look for programs that are more suitable. After doing this, you may find you don't even need to retake the GRE. 

And to clarify (because I know my posts are a little confusing), I don't think your profile is weak by any means. I just think you have one weak component, but it could be offset by a strength (e.g., research experience). If you're looking to apply to schools outside the top-10 and apply to at least say 10 schools, I'm 100% confident you'll get into at least one program. But it's hard to crack that top-10.

[huxlb]

@footballman2399 Thank you so much for the advice! I'm most likely taking Duke and CMU off the list, and having UNC and NC State as my only reaches. My top choice at this point is Virginia Tech though. I'm fairly confident that I'll be able to improve my GRE score over the summer, since I didn't really study for taking the GRE the first time. What range of a GRE score do you think would be necessary to be competitive?

[StatsG0d]

41 minutes ago, huxlb said:

@footballman2399 Thank you so much for the advice! I'm most likely taking Duke and CMU off the list, and having UNC and NC State as my only reaches. My top choice at this point is Virginia Tech though. I'm fairly confident that I'll be able to improve my GRE score over the summer, since I didn't really study for taking the GRE the first time. What range of a GRE score do you think would be necessary to be competitive?

I definitely think you're already a competitive applicant, it's just that the writing score is a little week. For most programs outside the top-10 you've got a very good shot already. I just think that improving the writing score to even a 4.0 would be good because a.) it's easily improvable and b.) it's the only true weak spot on your profile.

[huxlb]

Thank you! Do you think the quantitative score is high enough already, or would it be better to try to raise it?

[biostat43]

If you are really interested in Duke, I would encourage you to still apply. It seems that you have impressive research experience, which Duke certainly looks for. I am an applicant myself this season, but I have been admitted to Duke's PhD Statistical Science program with a profile that is fairly similar to yours. Best of luck!

[huxlb]

@biostat43 What was your profile like (GPA, GRE scores, etc), if you don't mind me asking?

[biostat43]

I'll be sure to make a detailed post using the standard template when I finish hearing back from the rest of the schools on my list. For now though, my GRE was 163 V, 165 Q, and 4.5 W, and I've had a few years of research experience in my undergrad (no publications yet). Probably the main difference is GPA (I have a 4.0). Anyways, I just recommend to not immediately rule yourself out for Duke and consider applying if you are very interested

[huxlb]

@biostat43 Yeah, since you have a 4.0, I feel like you definitely have a more competitive profile than me haha

[Bayesian1701]

@huxlb get Magoosh and try their advanced quant plan.  Maybe buy the official GRE writing practice to see if you can raise it a little.  My first practice test I a 164 and I was able to raise it to a 167 with a lot of studying over three months.  I would wait to study for the GRE until the summer and focus on grades instead.  If you are in any upper level math/stats classes try to get As in all of them.   A solid semester right before applications could really help you out. 

[Stat Assistant Professor]

Fwiw, UFlorida has just hired a bunch of new faculty and is making a concerted effort to diversify its expertise. The new chair of the department is a biostatistics guy, and we just hired new faculty in the areas of computational statistics, spatial statistics, and machine learning. Also, only one person graduating from UF this year wrote their thesis on MCMC. The rest of the theses are on a variety of topics, including vector autoregressive models, graphical models, network models, multiple testing, Bayesian nonparametrics, machine learning, and high-dimensional statistics. 

I'm not sure if you are considering UF, but it is a very Bayesian department (some frequentists), and I would say, within your reach given your profile. It has traditionally been quite theoretical but seems to be diversifying so it has strong faculty in more applied areas as well.

Edited February 20, 2018 by Applied Math to Stat

[huxlb]

@Applied Math to Stat Do you go to UF? I'm definitely considering them, and would love to know more about the program!

[Stat Assistant Professor]

13 minutes ago, huxlb said:

@Applied Math to Stat Do you go to UF? I'm definitely considering them, and would love to know more about the program!

Yes, and feel free to PM me if you have any questions. As mentioned above, UF is world-renowned in the area of MCMC, but many students do not focus on that (myself included). With the new chair, the dept is definitely diversifying, and there are likely to be lots of opportunities and changes in the future (e.g. there is talk of restructuring the program and the curriculum).

I myself got postdoc interviews at some top schools, so if you play your cards right, you can definitely succeed here.

[GoPackGo89]

32 minutes ago, Applied Math to Stat said:

Yes, and feel free to PM me if you have any questions. As mentioned above, UF is world-renowned in the area of MCMC, but many students do not focus on that (myself included). With the new chair, the dept is definitely diversifying, and there are likely to be lots of opportunities and changes in the future (e.g. there is talk of restructuring the program and the curriculum).

I myself got postdoc interviews at some top schools, so if you play your cards right, you can definitely succeed here.

What happened to Florida (as in why isn't it a no-brainer that you could succeed there / I don't see a lot of mention on these forums)? Off the top of my head it produced Brian Caffo, Brent Coull, it has or had James Booth and Casella. I always thought of UF as a great school for statistics but maybe I am biased because growing up in Florida the smart kids went to UF for undergrad.

Edited February 21, 2018 by GoPackGo89

[Stat Assistant Professor]

36 minutes ago, GoPackGo89 said:

What happened to Florida (as in why isn't it a no-brainer that you could succeed there / I don't see a lot of mention on these forums)? Off the top of my head it produced Brian Caffo, Brent Coull, it has or had James Booth and Casella. I always thought of UF as a great school for statistics but maybe I am biased because growing up in Florida the smart kids went to UF for undergrad.

A number of famous faculty members retired, died, or left to go to other schools (e.g. Casella, Young, Agresti, ...), so it lost some of its "elite"-ness this past decade. In U.S. News and World Report,  UF Statistics was ranked #9 in 2008, but has since fallen in the ranks. However, I would not concern myself too much with rankings.  Nowadays, a postdoc is practically a requirement for an academic job, unless you have at least one paper in a top journal or went to Stanford or something like that. 

 

What's most important is that you do good research, preferably with an advisor who is active in your field and who is well connected in the bigger statistics community.

[DJ3Sigma]

Plus Florida has the nicest weather in the states and has one of the only natural disasters that will give you weeks to prepare.

[huxlb]

Update: I have changed my list of schools a bit, and have taken a GRE practice test where I received a 157V/168Q/5.0AW. Any input on how this could affect my competitiveness at these schools would be appreciated!

• UNC
• Virginia Tech
• Ohio State
• Rutgers
• University of Iowa
• Florida State
• University of Florida
• University of South Carolina
• University of Missouri

Also, most of the B's I received were over one semester where I was doing statistics research ~30 hours a week. Do you think that makes a difference/should I mention this in my SOP?

Edited February 26, 2018 by huxlb

[GoPackGo89]

54 minutes ago, huxlb said:

Update: I have changed my list of schools a bit, and have taken a GRE practice test where I received a 157V/168Q/5.0AW. Any input on how this could affect my competitiveness at these schools would be appreciated!

• UNC
• Virginia Tech
• Ohio State
• Rutgers
• University of Iowa
• Florida State
• University of Florida
• University of South Carolina
• University of Missouri

Also, most of the B's I received were over one semester where I was doing statistics research ~30 hours a week. Do you think that makes a difference/should I mention this in my SOP?

One or more of your letter writers would also be good to have explain away those B's. I hope that is not your final list of schools but it seems you would get at least a couple offers from those nine schools you just listed.

[huxlb]

What about my chances for masters programs? I am thinking about just applying for masters programs first, to see if I can adjust to a program first. Any feedback would be appreciated!

[bayessays]

You would probably get into any master's program, especially if you got the 168Q. Unless you are fabulously wealthy, I'd suggest just going to a PhD program with the option of earning a master's if you drop out.