Columbia vs. UChicago vs. Duke for Stats PhD

[hopesandprayers]

I can't decide which one to go to, and I obviously need to by April 15th.  I feel like Columbia is the best fit research wise (I'm interested in applied probability/stochastic differential equations and Columbia has tonnes of probabilists), Chicago also has great people and is better if I change my mind and want to do more stats instead of probability (or for more theoretical probability - pure theory doesn't seem like a smart move career wise though), while Duke also has fantastic faculty and (this is somewhat important I suppose) seemed to have by a million miles the happiest grad students.

Any advice from those who have already made their decisions (or have already started/finished a program at one of these places)?

[Stat Assistant Professor]

Congratulations on your excellent acceptances. You can't really go wrong with those choices, but I would say you should strongly consider Duke if you could see yourself being really into Bayesian statistics and computational stuff. There is definitely theoretical research going on there as well (particularly for Bayesian nonparametrics), but there is also a big focus on Bayesian methodology and addressing computational challenges at Duke.

If you're agnostic about Bayesian vs. frequentist, then the other two might be better. It seems as though UChicago is the most theoretical of the ones on your list. Many talks I've attended by alumni and PhD students from UChicago seem to be solidly in theory (like proving risk bounds, attaining confidence intervals with the correct asymptotic coverage, etc.). So if you are solidly into theory, then that could be a good choice. Columbia seems to have a good balance of theoretical and applied/computational work, with some faculty who work a lot on statistical theory (e.g. Bodhisattva Sen) and others who work a lot on algorithms/methodology (e.g. David Blei). One other poster on this board was deciding between UPenn Wharton and Columbia Statistics, and they mentioned that they liked Columbia for being more applied than Wharton.

[hopesandprayers]

Definitely a Bayesian - wouldn't really be considering Duke otherwise, although it's definitely third on my list in terms of research I'd like to do.  I'm putting it up here because I loved it when I visited and they do still have an excellent department of course.  I lean toward Columbia because I feel like I should make this decision as much about research fit as possible, but I feel like I'm going to regret not going to Chicago because I absolutely loved the visit there too.  This is impossible for me to be honest.

[Stat Assistant Professor]

17 minutes ago, hopesandprayers said:

Definitely a Bayesian - wouldn't really be considering Duke otherwise, although it's definitely third on my list in terms of research I'd like to do.  I'm putting it up here because I loved it when I visited and they do still have an excellent department of course.  I lean toward Columbia because I feel like I should make this decision as much about research fit as possible, but I feel like I'm going to regret not going to Chicago because I absolutely loved the visit there too.  This is impossible for me to be honest.

I think Columbia has more prominent Bayesians (such as David Blei and Andrew Gelman) than UChicago. It seems like it is a good idea to trust your gut in most cases. UChicago is obviously a very elite program, but I would make sure they have enough faculty working on things you are personally interested in. Quality of life (e.g. living in Chicago vs. NYC) should also definitely factor into your decision. I would also add that there are other posters on this board who have turned down offers from elite programs such as Harvard, UPenn Wharton, etc. in favor of Columbia for reasons similar to yours (namely, a preference for the type of work done in the Columbia Statistics Department).

Edited April 12, 2019 by Stat PhD Now Postdoc

[hopesandprayers]

I would argue Columbia's stats dept is nearly as elite as Chicago's, perhaps not quite but somewhat close.  Some lists even rank it above Chicago - not on US new of course but that seems to be an average of scores given by faculty only.  I'm trying to make the "eliteness" of a program or its prestige mean as little as possible to me, but of course it still impacts my feelings about the decision (and I've clearly looked it up, lmao).  I think the quality of life in Chicago might be better due to the lower living costs and a more studious atmosphere (at least I got that vibe - and Chicago's campus is the best thing ever), but if we compare Hyde Park to Morningside Heights I'd say the latter is somewhat safer.

[EgoClaudius]

Congrats on your offers! It seems like the three depts are similarly appealing to you research-wise at this stage, so, if I were you, I’d focus on the quality of life for a PhD student in each dept. What are the funding packages at each dept? What do these become when adjusted for the cost of living in each city? What are the RA/TA responsibilities associated with each offer?

More generally, what college/city environment would you perform best in as a PhD student? When you visited, did students seem happier at one university than another/is the attrition rate lower at one dept than another, etc?

[Stat Assistant Professor]

17 hours ago, hopesandprayers said:

I would argue Columbia's stats dept is nearly as elite as Chicago's, perhaps not quite but somewhat close.  Some lists even rank it above Chicago - not on US new of course but that seems to be an average of scores given by faculty only.  I'm trying to make the "eliteness" of a program or its prestige mean as little as possible to me, but of course it still impacts my feelings about the decision (and I've clearly looked it up, lmao).  I think the quality of life in Chicago might be better due to the lower living costs and a more studious atmosphere (at least I got that vibe - and Chicago's campus is the best thing ever), but if we compare Hyde Park to Morningside Heights I'd say the latter is somewhat safer.

Columbia U also has a very elite Statistics PhD program for sure (as does Duke). I think for Bayesian statistics specifically, Duke and Columbia are stronger than UChicago. When comparing Columbia with UChicago, there are more prominent Bayesian statisticians at Columbia. If you know that you are strongly in the Bayesian camp, this is something that should factor into your decision for choice of program.

Edited April 13, 2019 by Stat PhD Now Postdoc

[penguinbombs]

It sounds like Columbia is the best research fit for you, and some of the other factors (UChicago's slight prestige edge within Statistics, Duke's happy students) are muddying the waters. But within this tier, differences in prestige are negligible, and I'm sure you'll be able to be happy at any of the three as well. Since Columbia is strong in MULTIPLE of your interests, I say go with that!

For what it's worth, I was admitted to these three schools as well. After the initial shock of getting into UChicago, I quickly dismissed it because it actually wasn't a great research match for me. Columbia was a great research match and was definitely the hardest school for me to turn down. I know how agonizing all this can be. If it helps at all, somebody posted recently about turning down Harvard and several other programs for Columbia. Best of luck with whichever school you choose!

[Spaghettini Plot]

Research fit is likely going to contribute most to your happiness later in a PhD.  Though the purchasing power of Duke’s stipend in Durham may also be keeping their students happy.  I think the slight differences in reputation between a place like U Chicago and Columbia would be overcome by the difference in where you feel you can do your best research.  

You seem to talk about Columbia in a way that you’ll be happy there and able to do your best work.  I say go for it! 

[hopesandprayers]

I just committed to Columbia, so thank you all for helping me get to the decision I was clearly trying to convince myself of from the get go