fujigala

@bayessays What I'm referring to specifically is maybe working with another professor outside of the department of biostatistics. For example, I know this is done in my statistics department where students will work with professors in other disciplines in addition to their own. But I don't know how acceptable it would be looked upon to do your dissertation in something completely unrelated to biostatistics. However, I do know that many master's students in my department will go onto PhDs in the business school, where professors still do highly theoretical research that is not directly related to business.

TLDR: Does it make sense to apply for PhDs in Biostats/public health even if I want to do research in something not really related (machine learning/neural network theory), on the basis that admissions are less competitive and it would be advantageous to be at a better school, and one can still do research in such an unrelated area? I'm currently a master's student in statistics looking to apply for PhD's in statistics next year to do research in machine learning theory (e.g., equivariant neural networks). My advisor in statistics recently came up with the suggestion to apply for PhD's in biostatistics/public health while still pursuing the same interests in machine learning theory (not really related to biostatistics/public health), since the admissions are apparently much easier. I would much appreciate any thoughts on the following: In a different field such as biostats/public health, is it still possible to do highly theoretical research in machine learning theory that is not really related to biostatistics or public health? My advisor seemed to think this was the case. If so, are there different ways to prepare for such programs? For example, for PhDs in statistics, taking a lot of heavy math courses is recommended. Is applying to such programs that are not exactly your primary interest advised or advised against? Biostatistics is one of my secondary interests, especially if it is hard to find careers in machine learning theory. Thanks in advance

My advisor has said something similar. He advises a lot of his master's advisees in statistics who are interested in statistics to apply for other less conspicuous programs such as operations research, management science, biostatistics, etc. with the idea that admissions are easier, but you can still take the same classes you would as a statistics major.

.

I am in the same position as you (OP). I chose Chicago over Duke. I had similar concerns to yours, with Duke MSS grads making ~100-110k and the majority of Chicago grads going on to PhDs. But Chicago was ~20k less per year, more prestigious, better PhD ranking, seemingly better overall school culture, and also much closer to where I live. I recall reading the Chicago statistics course catalog, and I think some of the core courses do involve more applied projects involving data analysis, etc. (not completely sure that they were core courses, but I think I saw this for some courses). You may want to check the core course descriptions in the course catalog yourself. Chicago does offer a really good variety of courses though, and I think you should be able to find some that will be more useful and applied than others (just a hunch, since there are so many courses, not 100% sure though). Also, worth noting that the thesis option at Duke is one of two required options, meaning you MUST do one in order to get the degree. The other is like doing two projects or something, which I would assume is significant in terms of time and effort. Keep in mind that at Chicago, your thesis, plus the required year and one project of consulting experience, could be potentially appealing to employers, to some degree, maybe (or not). At Duke, I do not believe any consulting experience outside of your thesis option is required. I also read on reddit some bad things about the overall school culture at Duke. About 25-50% of the people on there had really negative things to say about it, such as the culture being really cliquey, superficial, having some really unpleasant and rigid social hierarchy, difficult to make friends/date if you're not an athlete or frat/sorority member, etc. Just look up "Duke culture" or "Duke social hierarchy" or something on reddit. The remainder of the posters, ~50-75% said their experience was either mostly manageable and fine, or great, and such negative reviewers were probably people whose own problems were preventing them from enjoying their experience there. But there were enough complaints to be concerning to me, at least. On the other hand, I literally did not see any such complaints about Chicago on reddit, and most people rave about the culture there being great, not competitive, really stimulating, and the two students at Chicago (whom I've never met before) that I reached out to recently were extremely friendly. Most of this is from reddit, though, so take it with a grain of salt. What helped me make my final decision was just listing out all of the pros and cons of Chicago and Duke. Chicago really seems to me to have a lot more points in its favor. Even if you live near Duke and not near Chicago, I would still probably choose Chicago.

For MS programs with no research requirement, such as Stanford, do you think it is necessary to look into the research at the program and mention in the SOP which professors one would be most interested in doing research for? I am having trouble with this, since my interests are more applied, in biostatistics, finance, etc., and a lot of the research in statistics is heavily theoretical, i.e., developing new methodology, etc. Would it just be acceptable to state these areas of interest? My ultimate goal is just to work after I get my MS, maybe do some research over the summer. Any insight you may have into this would be very helpful. Thanks.

Thanks for the reply and info. Not sure if I want to commit 4 years to a PhD yet.

I am looking to apply to either stats MS programs or data science MS programs, but I need more information about both in order to decide which to apply to, and so I have a few questions: In your opinion, do stats MS programs provide enough training in applied and technical skills to work as a statistician or data scientist in industry? In looking at the curricula of many stat MS programs, it seems there is a heavy focus on theory and not as much focus on actually preparing one to work as a data scientist. E.g., not very much focus on how to actually apply machine learning models practically or much focus on computer science skills such as data visualization, software engineering, data mining, databases, cloud computing, etc. Is this perception wrong? Do many stat MS programs actually provide these types of skills within stat courses? Or do most stat MS grads pick up these skills on their own? It is somewhat difficult to tell just by looking at the program curricula. I would rather not have to learn these skills on my own, if possible. I am also wondering if data science MS programs would leave me as a good candidate for PhD programs in stats? Looking at some data science curricula, some or all of the courses are not offered in the stat department but offered in a separate data science department. Therefore I am wondering if stat PhD programs would view these courses as watered down and it would leave me as not a very strong stat PhD candidate. Would it be different if as a data science MS student you took some courses in the stat department? For example, Harvard's data science MS allows up to 4 stat electives. My primary goal with an MS program is to be prepared to work in industry as a statistician or data scientist, although I do enjoy math. My secondary goal is for the MS program to leave me as a good candidate for stats PhD programs as I am considering pursuing a PhD at some point but very well may not. Any insight you may have regarding either question or relevant to this topic at all would be very helpful. Thanks

Thanks for the reply

Thanks for the insight 1. Sufficient even for top 10/ivy league ms programs? 3. Not sure if I want to get a PhD at this point.

Hello, I just have a few miscellaneous questions relating to MS programs in stats/data science: 1. Do you think my GRE score of V: 165 Q: 165 is good enough to get into top 10 schools? What about Ivy League schools? I have seen the average quant scores at these types of schools to be 167+. I am considering retaking it because on practice tests, I did better on the quant section (167, 168), but worse on the verbal section (163, 162). I assume the rest of my application would give me a decent shot at top 10 schools. 2. Are lower ranked (based on USNWR rankings) Ivy League schools (and other schools with overall prestigious reputations) very competitive to get into compared to other schools with similar USNWR rankings but less of an overall reputation (e.g. Yale stats vs. UC Davis stats, both #31 on USNWR rankings)? I think having a degree from a prestigious school would help my chances of employment, but I don't want to apply to all Ivy League schools and not get in anywhere. 3. I am considering applying to MS programs in data science, but I am wondering whether an MS in data science might not make me as competitive for stat PhD programs as someone with an MS in stats. Is this a valid concern? Does it depend on how many stat courses you take as a data science MS major? My undergraduate degree was in physics, if that is relevant at all. I would really appreciate any insight you may have on any or all of these questions, thanks!

I got V: 165 and Q: 165. Do you think a higher Q score would significantly help my chances of getting into top 10/Ivy League schools such as Harvard, Chicago, MIT, etc.? I am considering retaking it since I got V: 162 Q:168 and V:163 Q:167 on my 2 practice tests

Thanks for all responses, this is quite helpful

I'm wondering if I could get some advice on what tiers of schools I should apply for, given my background? I am going to apply for an MS in stats (Fall 2021). Demographics: Biracial domestic male (Half-asian, half-white) Undergrad: UIUC (big ten, very strong physics program)~#37 rank in stats. Major: Physics GPA: 3.82 overall, 3.76 major GPA. Got a B+ freshman year in Calc 3, all B+/B's one semester junior year when I took 5 math/phys classes, otherwise all A+/A/A- Coursework: Calc 1, 2, 3 (A, A-, B+), Diff EQ (A-), Linear Alg. (B+), Discrete Math (B+), Real Analysis (A), Statistics and Probability I (A), Probability Theory (A+). A+/A in all CS courses (Intro CS, CS for science/eng., data structures). Notes: I took 3 years off in the middle of my degree taking part-time jobs+easy classes to deal with my depression after a death in the family. Research: One REU in physics, learning and applying machine learning methods to a scientific problem in biology (PI I worked for was in physics). Letters: I have an very strong letter from a PI from a physics REU where I applied machine learning methods to scientific research (he said I got more done than some grad students, told me I did a terrific job), one from a part-time tutoring job, and two from upper level STEM course profs in which I got A+ and A (Senior Quantum I and Intro to Real Analysis). GRE: Haven't taken the GRE yet, but on my first practice test I got V:157, Q:162. Misc.: Scored 2/120 on the 2019 Putnam exam, placing in top half of participants. I am wondering what tiers of schools would be best to apply for? Do you think I have a good chance at UIUC and other big schools in the top 50? Should I include some below the top 50? Do you think I have any chance at top 10 schools like Harvard, Berkeley, Stanford, etc.? I would really appreciate any advice on this or any related issue! Thanks!