Stat PhD: OSU vs. Rutgers

[statff]

Hello, I have been accepted into OSU(Ohio State University) and Rutgers. Though I'm currently waitlisted on several programs, I guess it would be difficult to get off the waitlist, so anyway I would like to make my decision between the two programs. Both programs are ranked similarly(OSU 24th, Rutgers 26th USNEWS).

I'm an international student and I have no concrete plan(industry or academia) after graduation yet. My research interests are mainly on high-dimensional statistics and my Master's thesis was about penalized classification problem under high-dimensional settings. Research fit is very important to me, but I understand that things could always change and saw many PhD students end up doing something completely different from what they originally had in minds.

As far as I have searched, OSU is slightly more inclined toward industry than academia, and many professors are working on Bayesian or spatial statistics, which I know very little about. Department size is relatively large, so there are many PhD students and professors of wide range of research interests including high-dimensional statistics. Coursework seems to be very intensive, including two qual exams, candidacy exam, and 80 credit hours. OSU also sent me some catalogue about its recent graduates' job placement records, which seems very solid to me. Columbus, where OSU is located, seems to be very nice place to live in too.    

On the other hand, I don't know much about Rutgers. Department size seems very small. Though there are as many number(30) of professors as in OSU, total number of PhD students are less than half. Coursework is relatively loose(just one qual exam). Total of 72 credit hours are required, but 24 or more of them are independent research or thesis related. Personally, I think Rutgers has more prestigious or reputable professors than OSU. There are many professors who are actively researching on high-dimensional statistics, including Prof. Cun-Hui Zhang. Rutgers also recently hired two professors whose main interests involves high-dimensional data analysis, which was the main reason I have applied for. However, I absolutely know nothing about its job placement records. In addition, living in New Brunswick is much more expensive. I get more stipends from Rutgers, but considering the high cost of living at NJ overall living conditions would be worse.     

I have hard time making my decision now. Any comments or help would be really appreciated. Thank you!

[Stat Assistant Professor]

I think Rutgers is a great department, with lots of great faculty (Zhang, Tan, Strawderman, etc.)! I interviewed with them for a job, and they do indeed have a lot of great faculty working on high-dimensional statistics. Some of the more nascent areas of research these days are high-dimensional generalized additive models, and constructing asymptotically valid confidence intervals for high-dimensional problems (when the 'classical' limit theory fails), and those were some of the projects they would have had me working on if I had accepted their offer. 

Rutgers strikes me as a more theoretical dept than OSU. If you are certain that you are *very* interested in statistical theory for high-dimensional stats (particularly of the frequentist flavor), then Rutgers is the better option, IMO. I asked Rutgers about their PhD graduates, and it seems as though most of their graduates go into industry in the NJ/NY metro area in data science or big pharma.

[statff]

@Applied Math to Stat Thank you for your reply! I've never thought someone would even have interviewed with them. Anyway, it was really helpful. 

[statff]

6 hours ago, Applied Math to Stat said:

I think Rutgers is a great department, with lots of great faculty (Zhang, Tan, Strawderman, etc.)! I interviewed with them for a job, and they do indeed have a lot of great faculty working on high-dimensional statistics. Some of the more nascent areas of research these days are high-dimensional generalized additive models, and constructing asymptotically valid confidence intervals for high-dimensional problems (when the 'classical' limit theory fails), and those were some of the projects they would have had me working on if I had accepted their offer. 

Rutgers strikes me as a more theoretical dept than OSU. If you are certain that you are *very* interested in statistical theory for high-dimensional stats (particularly of the frequentist flavor), then Rutgers is the better option, IMO. I asked Rutgers about their PhD graduates, and it seems as though most of their graduates go into industry in the NJ/NY metro area in data science or big pharma.

Thank you for your reply! I've never thought someone would even have interviewed with them. Anyway, it was really helpful.