kowtown

I disagree with nathany's post. I doubt you need the dozen or so computer science courses that constitute a major. You might want to take a handful of courses including introductory programming (if you're not proficient in C or C++), data structures, algorithms, etc. You've had more than enough applied stat courses. You probably don't need another major at all, never mind a degree, since admissions committees only care about what courses you took and how well you did in them. (Rarely does one return to undergrad after obtaining an advanced degree, but if you have the time and can get some sort of financial support for pursuing a second bachelor's, you may want to consider it.) What you need, as you seem to already know, is a good foundation in mathematics and a year sequence in mathematical statistics, which I'm sure you can get at U. Kansas. I originally majored in computer science. I didn't finish the degree at the time but I was just an elective or two from completing the major. I left with about the same math background you have now: two semesters of calculus and a basic linear algebra course (as well as discrete math). I returned to school a number of years later and completed an applied math major in two years, taking multivariate calc, two semesters of advanced calculus (real analysis), numerical linear algebra (to both brush up on linear algebra and introduce myself to numerical methods), a year sequence in applied statistics, and a year sequence in mathematical statistics. Of course I had some additional courses, like differential equations and stochastic processes, but these were more to just reinforce concepts in calculus and probability theory, and I would consider these optional. It may be premature to say so, but I feel this background prepared me very well for my first semester in a statistics PhD program. I regret not taking a more advanced theoretical linear algebra course, but I won't know until next semester how much that would have helped. I would recommend you take one since large data sets and visualization will each require vector and matrix manipulation and good grounding in associated theory. Of course one key token in the above paragraph is "PhD." You may not need to take all of the more theoretical courses like real analysis, mathematical statistics, or theoretical linear algebra, but taking at least a subset will make you a better thinker and improve your chances of getting into a good program. The program that you eventually want to get into will dictate what courses you actually take. I think I've said this before, do look at graduate programs you're interested in and see what they require and recommend, and do talk to professors and professionals in the field. Academics love to give advice! Also, once you begin your studies, you yourself will have a better idea of what you're doing and want to do. Best of luck.

As the field is interdisciplinary by nature I'd say take a mix of courses from all three, and select a major that allows you to do that. I would lean towards mathematics because not only would you acquire a set of skills but you'd become a better thinker. So perhaps applied or computational mathematics, either of which should allow you to take electives in computer science and statistics. Of course a pure math degree wouldn't hurt either. The rigor of upper-level math courses would likely impress admissions committees more than undergraduate computer science or statistics courses. If someone can grasp real analysis or abstract algebra, they can grasp anything. (Why take a watered-down math course? Though you may find it boring now, an advanced calculus or real analysis course is just good for you.) That should suggest your best course of action: Take a look at machine learning programs at universities you'd be interested in attending and see what they require/request of applicants. Definitely take a look at Carnegie Mellon--they have a department of Machine Learning in their school of computer science. But since you're only a sophomore, you still have a lot of time to figure things out. Best of luck.

I've been considering NCSU and Penn State. I have no idea how often students get involved with the various research institutes in the triangle, but it does happen. I think it's a legitimate reason to want to be down there. I hadn't thought about taking courses at the other universities. NCSU is a little farther from Duke and UNC than they are from each other, but they're all reasonable commutes. I can't speak for UNC but NCSU offers "industrial traineeships" where your stipend comes from working at a local company for twenty hours per week, and I would imagine you could set something up regardless. You would certainly have a variety of ways to spend your summers--while staying local--between working and participating in a program offered at one of the institutes. At Penn State you'd probably be a TA most of your time there, especially since English doesn't seem to be a problem. There aren't many businesses, so if you're interested in obtaining work experience you'd probably be spending summers in Pittsburgh, New Jersey, DC, the Triangle, or some other part of the country. In response to your other post, Pittsburgh could be a daytrip, while Philly, NYC, and DC aren't too far for a weekend getaway, when you do have the time. If you're into (or could get into) cycling, Centre County is supposedly a great area for day tours. I suppose I've been lucky in that most of the departments I've visited have been friendly, but other students visiting PSU with me noted that they're especially friendly. They're also pretty strong in just about every discipline, though I imagine UNC is too. Penn State seems like an awesome place to become absorbed in academia. PSU's department happens to be the best place for my current interests. That being said, while I would love to get back into fishing, hiking, and biking, I've become more of a city person, and unfortunately University Park/State College just doesn't have a lot to offer my significant other. The Triangle isn't exactly a cultural mecca, but it does have a lot to offer among the three cities. But I prefer compact Northern cities to sprawling Southern ones, so I'm strongly considering an offer from Pitt. I worry the department may be too small, but there exists the opportunity for collaboration with other departments, as well as with CMU. If I can't see myself being happy at one of these places for five years, I may just take a funded master's from South Carolina and reapply once my interests are more evolved. If anyone is on the waiting list of the Statistics departments at Penn State, NCSU, Pitt, or South Carolina, a spot could be opening up at two or three of them by tomorrow, and almost certainly by Monday.

I wouldn't think so. I emailed Dr. Davis and he let me know right away that I had not made the list.

I don't know much about LA but I know Pittsburgh has weathered the recession quite well and California has not!

Agreed. Even if you're going into industry, with a PhD you're applying for a research position, and they will know UMN. However, CMU is obviously a great program as well, and Pittsburgh is a great city and an important hub of research (at least for statistics and computer science--I can't speak for applied math but I imagine there are opportunities). You really need to focus on their respective strengths in relation to your interests.

I hope I'm not too late. While some programs to which I was accepted offered travel expenses up front, others obliged once I inquired, at least putting me up for the night. I would imagine this would be a possibility at the PhD program, but maybe not at the MS program unless they're serious about recruiting--I'd think it would be in their interest to fly in a candidate who got accepted to a respectable PhD program if they want to improve their own. Maybe you can use the other offer as leverage. As for deciding between cities, it sounds like at school B you would actually be able to afford daycare between the secured funding at the school and the job opportunities in the surrounding area. School A could mean a longer time to degree completion if you can't secure funding for your research or some courses aren't offered regularly. Otherwise I agree with the above poster that you will probably be too busy to really enjoy the city.

Sorry, yes. B/I/G is Business/Industry/Government.

phd.org allows you to use your own criteria to rank programs. The data concerning quality of faculty and program effectiveness do indeed come from the 1993 NRC survey. The quality of faculty rating is what determines the rankings. US News uses a similar methodology for most fields. I agree that 17-year-old NRC rankings are highly inaccurate. Most faculties have undergone significant rotation. UMD has undoubtably become a significant force. However, they are useful to gauge things like historic reputation and visibility. And it's all anyone has had to go by outside of the US News rankings of the top ten. I believe US News only ranks the top ten. If it were only a matter of paying for access, a lot of us would have done it--a subscription is probably cheaper than an application fee. They posted an article saying that this year they are ranking statistics departments for the first time. These rankings come out April 15th, just after most of us have made our decisions. I do think NCSU is a good program. They have excellent placement, in both academia and B/I/G. It's probably just because NCSU is a much larger program (the largest in fact), but I see a lot more NCSU alums on faculty lists. And if one's focus is applied, I think they are hard to beat due to their position in the research triangle and industry connections. You can even receive your stipend through an internship rather than grading papers.

NCSU is only two spots below Yale in the 1993 NRC rankings, with 3.54 Q-points vs. 3.62, a difference of less than one-tenth, probably within the margin of error. Are you going by some other rankings?

The Grad School Rankings Are Coming Soon So basically they're using the same methodology that they used for the math specialties rankings in 2006 and 2008 and that the NRC used for the 93Q ratings. Does anyone care to make their predictions for the top 20 based on where they've been accepted and rejected? Or, better yet, who has an advanced copy?! I'm not rankings obsessed, but there does seem to be a correlation between rankings and placement!

No problem. The 1993 NRC rankings Note that I believe the NRC actually didn't actually rank departments, but this TAMU statistics professor compiled the list based on their ratings. You will observe also that it includes Biostat programs. For some schools which have both, it's not immediately clear which they might be referring to. A post listing the 2006 US News rankings The 2008 US News rankings

Indeed. You can even transfer between the Math Stat and AMSC programs and keep your funding as they're both run by the same department.

KatieM, I applied to GWU and considered Maryland (College Park) and UMBC. GWU had a respectable ranking at #32 (with 2.91) in the 1993 NRC rankings for Statistics programs, while Maryland came in at #18 (with 3.97) and UMBC at 115 (with 1.69) in the Mathematics rankings. A professor I consulted had good things to say about Baltimore County for statistics but didn't know much about College Park's program. I get the impression that the latter is up-and-coming. It certainly has a top-notch mathematics program, and the department shares resources. (After putting that in writing I wish I had applied.) Outside of the top ten it's definitely hard to get anything beyond anecdotes. Did you try "Maryland" or "College Park" in the search feature? I thought I saw it in someone's profile or a post. Have you heard anything from GWU? I'm going to be on the East Coast next week for spring break and they're open, so I thought I might stop by, but I haven't gotten a decision yet myself. (It'd be nice to do it on their dime rather than my own!)

I'm sure you'll get a lot of responses recommending you take some much deserved and probably needed time off, but I can certainly understand the desire to be productive. I wish I knew where I'll even be during the summer. Depending on where I get into (Statistics, btw) and what position my girlfriend gets, I could me moving to an as-of-yet-unknown destination at the beginning, middle, or end of summer. But if I stay here I know I want to take or audit some courses to get ready. My university offers a number of graduate courses during the summer in both math and statistics, and I imagine the one I'll be attending in the fall does too. It sounds like yours doesn't, but since you're Applied, maybe there's a field of application you'd like to investigate, in which case an undergraduate course might be in order. And of course there's always work--saving up for the move, apartment, big city life, etc. I'd ask your professors. If your school doesn't offer anything you'd like to take, maybe you can do research or an independent study, or work for the department.

I have similar interests, and I too switched from econ to stats, though as an undergrad. I desired the PhD but I took a look at the short-term prospects and decided statistical consulting was better than economic consulting. I also became more interested in statistics itself than economics while taking probability and regression courses in preparation for econometrics. I just finished applying to a mix of master's and PhD programs, so I can offer some recommendations. My background is not as strong as yours but I'll let you know how it turns out. Hopefully the economy is doing better next season because right now it seems that there are very few spots for very many applicants. Your profile seems nearly ideal. I wouldn't shy away from the top programs, but apply broadly. You have stellar marks in the right math, statistics, and computer science courses. You also have real-world experience working with data sets. A field of application is usually desired, so I'd highlight some of your experiences in economics in your application as well. Definitely brush up on calculus, take some practice tests, and shoot for 80th percentile or better on the subject test. A general test quantitative score near 800 is expected. Even if you can't achieve these, though, the other areas of your application could make up for not-so-stellar test scores. Decide who can write you glowing letters of recommendation and reinitiate contact. At least two of these should be from math or stat professors if possible. (These may also be the best people to ask for advice.) I wouldn't worry about the lack of research experience too much--a group paper on random walks is better than nothing, and I don't think much meaningful research is done at the undergraduate level anyway. You haven't had sufficient exposure to statistics to decide on an area of research, but definitely do find some areas you're interested in and communicate them to the admissions committees. By this I mean theoretical areas, topics of statistics where you'd be happy developing new theories or methodologies, ideally with applications to the social sciences or similar. Many people choose statistics for the flexibility of applications, but remember you'll be a statistician first. When researching schools click on the faculty research link or the faculty profiles and google/wikipedia anything you're unfamiliar with or that sounds interesting. Bayesians seem to be particularly interested in applications to the social sciences. Washington and Harvard are very good choices for social science applications, as are Cornell and CMU. Many members of the Cornell faculty have cross-appointments with the Social Statistics department of the ILR School (Industrial and Labor Relations), and the CMU Department of Statistics is housed in the College of Humanities and Social Sciences and prides itself on cross-disciplinary research. Both Larry Hedges and Bruce Spencer are Faculty Fellows at Northwestern's Institute for Policy Research in addition to being professors in the statistics department there. Other departments with members interested in social or behavioral science, education, or policy applications include U Pitt, Wharton (U Penn), Columbia, and George Washington. UC Santa Barbara offers an emphasis in Quantitative Methods in the Social Sciences. I'm sure there are more. You're at an advantage starting this early. The above are good places to start. The search can be exhausting, but probably worth it in the end. Begin working on your statement of purpose after you've found some areas of interest. You know you're applying to the right school when the "fit" paragraph comes easily. Good luck!

I just happened across admissions statistics for U. Penn's math program. If they have similar data for the statistics program I am not aware of it, but I imagine the numbers would be comparable. The pertinent information is:

I've applied to a number of schools and a few have told me I can log in to the same application page to check my status, others have given me a separate page and new login information, and, well, others just aren't so tech-savvy.

I am about to apply to a mix of MS and PhD programs in Statistics and am having trouble narrowing down my list of schools. As an undergraduate, my interests remain broad, though with a few stats courses under my belt and having looked into faculty research at schools I am considering, I have at least a basic idea of what I do and do not want to do. I am interested in both theoretical and applied statistics; I am particularly interested in applications in the social sciences. This has helped me narrow it down somewhat, but most schools have a sufficent number and diversity of faculty to cover most areas. At this point I am mainly interested in determining where I might be competitive. I can't afford to apply to many reach schools. I have sought counsel from my professors who have been a help, but they have limited experience sending undergraduates to programs. In general I feel they think I am more competitive than I really am. I am focusing on schools on the coasts; my partner will concurrently be applying to university staff positions, particularly in international programs and academic advising. I have a perfect record at my current (tier 4) undergrad, though at my previous (tier 1) undergrad, over 7 years ago, a less mature me had stopped attending classes and received a number of F's in two noncontiguous semesters. I am holding back my transcript until my current grades are posted. After this my GPA in all coursework in the last 7 years (in actuality, a year and a half) will be 4.0, my "last 60 hours" will be 3.2, and my overall undergraduate GPA will be 2.7. Math and stat courses include, with A's except in the one indicated: Discrete Math, Calculus III, Linear Algebra ("C"), Math Modeling, Mathematical Methods for Economics, Advanced Calculus I/II (rigorous), Numerical Linear Algebra, Intro to Probability and Statistics (calc-based), Intro to Probability Theory (calc-based with some proofs), and Statistical Methods and Models I/II (linear regression, ANOVA, and categorical data analysis). Additionally, on my old transcript I have 9 computer science courses with A's or B's, and an economics minor at my new school. Next semester I plan on Intro to Mathematical Statistics, Ordinary Differential Equations, and one of {Stochastic Processes, Abstract Algebra, Complex Analysis, Numerical Analysis}. (Any recommendations?) My GRE scores are 790 Q (92%), 710 V (98%), 5.0 AW (81%). My recommendations should be fairly strong, though I don't believe any of my professors are star researchers. I myself have no research experience. Any help would be appreciated. If you are hesitant to name specific programs I completely understand; feel free to list representative schools or ranges, or to PM. Thank you very much. In return I will post my schools in my profile once I've applied and keep everyone updated when decisions come in!

I grew up in and around a large city and got my license later in life, so this is something I can relate to. I would say that most college towns are pedestrian-friendly, since most students can't afford a car, but if you're looking for more culture, most large cities should have decent public transportation; some of my favorites are below. I think it would help if you specify Master's or PhD, but below are some schools that have at least some graduate mathematics program. I myself will be applying to Statistics master's programs in some of these very same cities. I would certainly second NYC. I've been by Columbia, NYU, and a CUNY or two and they're all easy to get to, especially NYU. The West Village has got to be one of the greatest places in the world to go to school (or at least to hang out in). It's also one of the most expensive to live, but Jersey City and Brooklyn are relatively inexpensive and easy to commute from. The Boston/Cambridge area is also very pedestrian-friendly. Everyone is super fit from walking all those hills! Boston is also expensive but that of course is true of any city so dense. There have got to be more than just those three schools to consider though. Try Boston College for a fourth. I've never been to Penn but I've been to Philly a few times and while I haven't utilized the public transportation very much, for the most part it too is pretty walkable. Temple and Drexel also have programs. Chicago, too, has other schools. Northwestern is not in the city proper, but it seems to be accessible by public transportation. Another large school is UIC. I've honestly never been a big fan of the city's transit system--the El's are kinda slow--but I wouldn't say it's hard to get around in. But it's no New York! Though I've been there, I can't personally speak for D.C. I've walked around but never used the public transportation, and I'm not terribly familiar with its schools, but Georgetown, GWU, and American University should all be good options if they have relevant programs. Finally I can recommend two other cities with great public transportation: Portland and Austin. In Portland, try Portland State University. In Austin, give UT a shot. I see someone's mentioned Berkeley. I'm not very familiar with CA, but it has so many schools--let's see some more! I've only been to San Diego, which very much seemed to be a motor vehicle town.

britten2, There is a forum for people who are very serious about PhD economics over at TestMagic: http://www.urch.com/forums/phd-economics/ Don't let them scare you about all the math they're taking though. The mathematical preparation borders on obsession. Consider their recommendations and look for common themes. Some people get in with just multivariate calc and linear algebra if their applications are otherwise strong. But most people would say to take, in addition, advanced calc/real analysis and mathematical statistics. Also particularly useful are differential equations and optimization/non-linear programming (possibly via a mathematical economics or operations research course). A more advanced analysis class would be a plus. From what I've heard, these should get you through the core courses. Don't worry too much if you don't get a chance to take them though, because you will likely go to math camp in the weeks before your program starts. Abstract Algebra and Number Theory have very limited application in economics, unless you know something that I don't. (I certainly don't know much about International. I was considering the PhD and have done a fair amount of research, but I'm not committed yet so I've decided to pursue a statistics master's.) They won't hurt though, because they'll demonstrate further evidence of proof-writing skills. But more analysis would probably be better. This is another good resource: Books to Study Before Going to Graduate School in Economics Definitely talk to your economics professors about applying. A lot of economics majors are pursuing private work or the MBA, and you should find that professors are very eager to talk about the PhD and can offer better recommendations than anyone else. Good luck.

I dropped out of a Tier 1 school 6+ years ago. My GPA started out stellar at a school with a reputation for grade deflation, but my non-major grades dropped over time until I finally stopped going to all my classes and received quite a few F's, killing my average. (From what I've read in forums like these, this need not be a death knell.) So I have a fair number of good Computer Science grades from there (halfway between a minor and a major) with the rest of it mostly irrelevant to Statistics, save a C in Linear Algebra, which is becoming a cause for some consternation! I completely turned around and with a renewed interest in formal education enrolled at the local Tier 4 as a math major. I've got a 4.0 through a full year and intend to keep it, probably for exactly 60 cumulative hours. (I also expect to do well on the GREs.) From this institution I've got most of what's usually recommended: [*] Calc III (multivariate) [*] Advanced Calc I (sequences, functions, continuity, differentiation, Riemann integrals) [*] Intro to Probability and Statistics (not very broad or deep, but with double integration) [*] Probability Theory (first bit of Wackerly/Mendenhall/Scheaffer) [*] Statistical Methods and Models I (applied linear models plus a SAS lab component) Additionally I'll be taking the following in the spring, after I've already applied. I wish I could've taken them sooner but they're only offered once a year and the prereqs precluded me from taking them last year. I'll probably mention them in my SoP. [*] Mathematical Statistics (the next bit of W/M/S) [*] Stochastic Processes I'll be applying this fall to graduate Statistics programs, and I'm undecided on what to take that semester. Tell me if it seems silly, but I'm worried about a perceived deficiency in Linear Algebra. I'm perfectly capable of brushing up on my own, but since I took the class many years ago and got a C, I'm a little worried about what the adcoms will think, since it's such an important base for everything else. Here are some options I have, the first two to improve that deficiency, the second two for some additional stats. MATH 2xx - Linear Algebra and Applications Matrix algebra and solutions of systems of linear equations, matrix inversion, determinants. Vector spaces, linear dependence, basis and dimension, subspaces. Inner products, Gram-Schmidt process. Linear transformations, matrices of a linear transformation. Eigenvalues and eigenvectors. Applications. Constructing and writing mathematical proofs. A transition between beginning calculus courses and upper-level mathematics courses. PRQ: (Multivariate Calculus) MATH 4xx - Numerical Linear Algebra Roundoff errors and computer arithmetic. Direct and iterative methods for solving linear systems; norms and condition numbers, iterative refinement. Linear least squares problems: the normal equations and QR approach for overdetermined systems. Numerical methods for eigenvalues: an introduction to the QR iteration. Extensive use of computers. PRQ: (Multivariate Calculus, Linear Algebra, and Computer Programming) STAT 4xx - Statistical Methods and Models II Continuation of STAT 4xx. Topics include factorial experiments: interactions, nested models, and randomized block designs. Categorical response data analysis: ordinal data, measures of association, Cochran-Mantel-Haenszel Test, logistic regression, and measures of agreement. PRQ: (Statistical Methods and Models I) STAT 4xx - Statistical Methods of Forecasting Introduction to forecasting including use of regression in forecasting; removal and estimation of trend and seasonality; exponential smoothing; stochastic time series models; stochastic difference equations; autoregressive, moving average, and mixed models; model identification and estimation; diagnostic checking; and the use of time series models in forecasting. PRQ: (Statistical Methods and Models I) I've already missed the boat on a more advanced linear algebra (LA) course. I'd rather not take a 200-level introduction to proof course my senior year, especially after taking Advanced Calc. At the same time, this is a proof-based LA course that strong applicants will have As in. Numerical Linear Algebra (NLA), OTOH, is somewhat interesting to me, reviews essential LA topics at the beginning of the course, and teaches Matlab. (Re: skibum's recommendations: The text, at least, covers algorithms for SVD, QR, and Cholesky--see "Preparation for Statistics MA?" below.) NLA has applications in statistical computing. It's not what I plan to study at a higher level, but with my CS background, who knows. It may also remediate my absence from the computing world. Numerical Analysis is an option for the spring, but I think that would be too late for an ancillary course to have an impact. Another stats course would be icing on the applied cake, but could help me secure another recommendation. LoRs were, unfortunately, an afterthought, but I should have two lined up soon. I don't really have a third professor that knows me very well as of right now, so one letter might have to come from a class I take this fall. Forecasting seems really interesting, but this is something I would like to take when I get where I'm going, and unfortunately it conflicts with Advanced Calc II (series and multivariate functions), which I'd prefer to have on my transcript this semester. Let me know if you disagree. The second methods and models course is less interesting, but could help me get a handle on what I want to study. Econometrics is also an option, though the course offered in the fall overlaps a great deal with the probability and regression I've already studied. The professor--who, though from the econ department, might give me a strong recommendation--does teach matrix theory in the course, but it's not in the course description. (Do adcoms even read descriptions or do they depend purely on title and what might be said in the SoP?) A more advanced class is taught in the spring, but, again, no one will care about the spring. I suppose I should also state that I have an interest in economics and initially intended the Statistics MS to be a stepping stone to the Economics PhD, but I am far from committed to the latter and now see the former as an end in its own right. That said, if I don't take one of these stats courses I could fit another econ course that could help me decide. My main objective is to get into a good Statistics program though. I plan on working advanced math and econ courses into my master's program as long as the PhD is a possibility. I ran the NLA course by someone in the stats department, and he seemed to think it was a decent idea, especially the Matlab part, but he didn't know much about placing students. The math advisor recommended Abstract Algebra I instead, since it's more rigorous, but it has almost no relevance beyond additional proof writing, which I think is fairly well covered by the more relevant (and more challenging, at least here) analysis courses. I realize this has become quite a long post and I appreciative you taking the time to read all the way through it. I hope it is of at least some interest to you, perhaps as an indication of what other people are taking or thinking or fretting about. Any comments would be greatly appreciated, even if they are about my current mental state. For recent posts on general preparation for the Statistics MA/MS, see Preparation for Statistics MA? and Would anyone accept me? P.S. I'm not sure where I'm applying yet, but I'm leaning towards east coast cities as far as location, ideally NYC (a region I used to reside in). If anyone could give me an idea of what level I'm competitive at, that'd be greatly appreciated too.