efh0888

Hey @footballman2399 what's your source on this? Are you a student there? Can't find anything about funding for MS students on their website!

Choose a different person for the last LOR if possible. Agreed, Magoosh is simply incredible. Shoot for 168+ on Q. Trust me, it's possible Not sure if research experience matters. Think math is probably more important. So yes, still take real analysis as it'll help once in a program. Read the thread below about choosing schools. I strongly agree with @TakeruK. Anecdotally, my first time around (long story) I applied only to reach and match schools in the top 20 and got great results. With no research experience I might add. Of course, YMMV.

Also love hearing from the experts, so if @cyberwulf and/or @biostat_prof would be willing to chime in that would be wonderful.

Bump... I've been looking into this again more recently. Rob Weiss at UCLA suggests focusing on math (https://faculty.biostat.ucla.edu/robweiss/preparation) which is what I'll probably do. But like @compscian said, I agree that gaining research skills is ultimately more important than coursework. I have a family so really need to keep working full-time as long as I can, so I don't have many options to get both the math and the research. So I've found a synchronous/online MS program in Mathematical Sciences housed in a combined Math/Stats department that actually looks very rigorous. It is light on course requirements, only requiring 2 core classes in matrix theory and mathematical statistics, and the rest is pretty much open (just needs advisor approval). They offer a bunch of stuff online on top of the core including real analysis, complex analysis, topology, numerical analysis, advanced probability and inference, statistical modeling, regression, DOE, etc. The best part is they also offer a 2-semester research course as well as a thesis option. Now for the big downside -- it's from a no-name state school. What are your thoughts on how much the prestige of the school would matter in admissions? For my situation, I'm hoping that adding the graduate math coursework plus relevant research experience from any school will put me in a solid position. But hey, I figure an "ensemble" of diverse opinions will perform better on average than just one, right guys?

With a few exceptions, I've been pretty skeptical of the value of masters degrees in data science/analytics/business analytics, favoring a degree in one of the pillar fields like statistics (not applied statistics) or computer science. The main drawback to this approach, of course, is the course requirements in areas that aren't that helpful for data science. However, at the PhD level, since research these days is highly interdisciplinary, especially in domains like machine learning and data visualization, I'm not so sure what to think. There aren't very many such programs yet. Some examples: http://csm.kennesaw.edu/datascience/ https://www.wpi.edu/academics/datascience/phd-program.html http://bas.utk.edu/academic-programs/phd/default.asp Also, I know of another state university currently developing one within the stats department in collaboration with the CS department. I'm very interested in statistics but worry it'd be too theoretical. I'm also interested in biostats but don't necessarily want to be pigeonholed into one application domain (even if just during the PhD). I would actually be most interested in CS if I had taken enough courses in undergrad to meet admission requirements. So I am curious about your thoughts on these types of PhD programs, particularly for someone who wants to work in industry.

Since you included things that won't matter like club involvement and internships but didn't mention your math background, thought I'd let you know that PhD level Finance involves a lot of proof-oriented mathematics. If you haven't taken this kind of math, you should start there. I'm not sure more finance experience, even in research, will help at all without the math.

That's encouraging actually. I took a look at some student profiles and you're absolutely right. In fact what draws me towards biostats is the focus on applications and methods over theory (while still taking a rigorous approach), so I'd much rather go for an MS in stats or biostats instead of pure math. Additionally, I'm sure there are more opportunities to work with truly big data sets and to get exposure to areas like machine learning.

@arima thanks, you make some good points. One thing I forgot to mention – of the ones I've been looking at, i.e. in-state options for financial reasons, the math degree allows for a thesis whereas the other 2 do not. I have a full-time job I love, and again for financial considerations, would almost surely do an MS part-time making it difficult to get research experience otherwise. How important do you think research experience is for biostat phd admissions? Is that a major factor to consider between the degrees? Also, what about factoring in schools like FSU and NC State where biostat is part of the stat department? Would they look down on a stat MS? FWIW in undergrad took linear algebra, calc sequence, diff eq, an intro to proofs type of course, stat theory 1 & 2, and a lot of stat methods courses. Also took grad level stochastic processes and advanced calc while in my first master's. Advanced calc was only 1 semester and was really tough but imagine it was light compared to a real analysis course. So I think I've got the chops for either math or stats. Also, not sure if it would matter for the choice here but my goal after the PhD would be industry.

Thoughts? Statistics - focus on data mining but includes a stat theory sequence (at the level of Casella and Berger I believe) Biostatistics - seems light in terms of theory but gives exposure to public health Math - not applied, mostly theory but includes some computational classes, and has a concentration in where you take a sequence measure theoretic probability and mathematical statistics

It's been noted many times in these forums that most of the Canadian schools offer funded MS degrees in all sorts of disciplines, including Stats and presumably Biostats.

NCSU also has a pretty thorough list, although it's probably not exhaustive: http://analytics.ncsu.edu/?page_id=4184

I'm not in a biostat PhD program, but I imagine at any top 10 program for any quantitative discipline, a quant GRE score >= 90th percentile would be necessary but not sufficient for admission, meaning they probably won't even look at the rest of your great application. See http://www.ets.org/s/gre/pdf/gre_guide_table1a.pdf for scores and their percentiles. I say re-take the GRE. Do everything you can to ace it. I recommend Magoosh. I'm a terrible test taker too and got a 95th percentile score no problem. If you can't, or more importantly, if you are unwilling to put in the time then you would not succeed in a top 10 PhD program anyway...

Agreed, the heavy focus on SAS at NCSU would be a negative from my perspective, too. This day and age it's all about open source tech like R, Python, Hadoop, etc. To clarify, for the Northwestern program, you mean the on campus one in Analytics offered by the College of Engineering? That one looks legit, not as sure about the online MS in Predictive Analytics. BTW, I wouldn't say I have a negative view of bootcamps, more just skeptical that it would be helpful for someone in your position. I could very well be wrong. Try finding and contacting alumni that went in with a similar background as yours. See if they think it helped with placement and prepared them well.

If you really want to be in a technical data science role as you mentioned, then I would recommend excluding business analytics degrees from consideration. Personally, I would do a MS in a more "pure" field like statistics or computer science, if you have the right academic background, and learning the data science specific skills through self learning, MOOCs, Kaggle competitions, and so on. The lone exception is the analytics program at NCSU, only because they're results, which are transparently posted online, speak for themselves. Additionally, the program is only 10 months, which lands you somewhere between a traditional MS and a bootcamp in terms of the time and financial commitment. FWIW my impression is that the bootcamp space seems like it's primarily designed for people with graduate degrees in quantitative areas who want to quickly transition into data science.

Do you mean MSU?

I HIGHLY recommend Magoosh. The best $99 I ever spent on test prep materials. Bumped from a 161 to 168 on Q in about 4 weeks.

I think you'd be fine in terms of GRE (sorry, no the better V won't help but you're Q is fine) and courses since you have all the math and programming reqs. I'm sure it'll be helpful that you've had exposure to writing proofs (via the Theory of Single Variable Calculus course). Real Analysis would probably be overkill for the MS level, especially if you have no plans to do a PhD afterwards. Anyone disagree? For schools, I'd say pick 1-2 schools in each tier, which makes 4-8 total. I believe they've been broken down by tier in various threads on here before. A special note on rankings: it's important to rank stats separate from biostats, which you did not do when saying Yale is 34th. They're probably more like 20-25 when removing biostats depts. Also I'd recommend you throw in a super safety school that you would have like 95% confidence you'd get an acceptance but make sure it's somewhere you'd actually be willing to attend if it was your only acceptance. That's obviously a personal choice.

It is: https://en.wikipedia.org/wiki/Terminal_degree

Fyi just to clarify terminology, "terminal" masters means they are designed to be a final degree, i.e. not as a stepping stone to a PhD. If you mean standalone masters, as in not a masters en route to a PhD, that's different.

I think Chicago requires the math gre, too, but I'm not sure. Seems like it's crazy competitive right now, because of all the hype around "big data" and "data science". I guess if you're gonna do a Ph.D., though, Statistics is a pretty awesome choice. On the career side, have you seen this article, http://fortune.com/2015/04/27/best-worst-graduate-degrees-jobs/ ? Any news on doing the thesis with ISU?

Cool, congrats. I'll probably apply in the not so distant future. Your background seems appropriate, including a solid math background and a touch of research experience... Two things worth noting: 1. You need to retake the GRE. Shoot for a 95%+ score in Q. I highly recommend Magoosh. Additionally, for many top programs, I think they will require the mathematics GRE, which I hear is tough, even having taken real analysis and measure theory. 2. Think about why you would want to do a Ph.D. and have really good reasons. Anecdotally, I went for mine in Industrial Engineering at a top 20 department and probably top 2-3 in Industrial Stats; it was miserable, because at the time, I wasn't passionate about the research, so I ended up dropping out. If you aren't in love with the research, I'd recommend finding something else to do.

Any results yet? I'm also looking into online M.S. degrees in Stats that are both affordable and mathematically rigorous; the ISU program is pretty much all I can find that fits both. NCSU just launched an online M.S. that looks promising. But it seems like they won't be offering the more advanced versions of the theoretical stats and linear models courses like you could take in person on campus. Bummer...

I can't say for sure, regarding how strict UCLA would be on this, but I think mathematically intensive economics courses, like micro and macro theory, econometrics, and so on, would be included. Assuming the lower level sequence to which you are referring is Calc 1-3 and Diff EQ, you would probably really want to take, all of the theorem-proof variety, a linear algebra class, a course or sequence in advanced calculus (i.e. lower level real analysis), and a calculus-based probability and mathematical statistics sequence. Lower level stat methods courses probably don't mean much but don't hurt obviously. Familiarity with software like SAS, R, Matlab, etc. would probably also help.

I should note, though, when you say for CS that, "it seems like you can't even think about admission to the top 5 without publications," I believe this is really true for any quant discipline. The competition at the top is fierce; I couldn't crack the top 5 in IEOR this year, and, of course, I had no pubs. ASU has a top 15-20 PhD program in IEOR, and is arguably the best program in Industrial Statistics (within IEOR), except for perhaps Georgia Tech. I got in with full funding and two good fellowships. This year there were only 15 admits with funding out of 150+ applicants. Maybe that gives some perspective, at least anecdotally?

Generally speaking, I think you are right. There are just a ton of people doing CS right now, so it's more competitive. One thing to note, if you are doing an MS, your job opportunities would probably be different coming from a IEOR or Stats background vs CS. At least that's what I would think...