Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Which undergrad comp sci courses would be helpful for stats?

  1. Feb 16, 2017 #1
    I'm two semesters away from graduating, after which i will get a masters degree in stats. My last semester is basically just one where i take courses I want while i wait for grad school to start, because i will have graduated after the semester after this current semester. Which COMPUTER SCIENCE courses should future data scientists, data analysts, and statisticians take to prepare for grad school and the work force?

    And while we're at it, if you know which graduate level courses would be most beneficial for this purpose (as i will concentrate in data science), i would greatly appreciate it if you could tell me those too. Thanks!
     
  2. jcsd
  3. Feb 16, 2017 #2
    Probably a course in SQL. Most big data that you'll need stats on end up in a database. Knowing how to write efficient queries is important.
     
  4. Feb 18, 2017 #3

    StoneTemplePython

    User Avatar
    Gold Member

    I'd suggest a course in Algorithms if you haven't already taken one. Understanding things like dynamic programming (as an algorithm design paradigm) and what makes a problem tractable vs intractable... is quite important.

    There are other courses tied in with machine learning that you may like. For example: Probabilistic Graphical Models are sometimes taught by comp sci departments, and sometimes by statisticians (e.g. Spiegelhalter is a co-author of a book in this space).
     
  5. Feb 21, 2017 #4
    Machine Learning is an obvious choice. Visualization is another possibility. Or just take something interesting for fun, why not. Theory of computing is a good course for honing your skills in constructing proofs and thinking abstractly, which I think is good for everyone. Algorithms may be a useful as well. Bioinformatics could be an option if you have the biology prerequisites since it is very stats heavy. You could also consider parallel programming. Scripting languages could be another good option.
     
  6. Feb 21, 2017 #5

    FactChecker

    User Avatar
    Science Advisor
    Gold Member

    What courses are offered that you have the prerequisites for? Can you program a computer at all? We could suggest a lot of courses that you might not be ready for.
     
  7. Feb 21, 2017 #6
    currently taking intro to programming ( again, since the uni doesn't accept my credits from the cc course of the same title). I have room to take one prereq next semester
     
  8. Feb 21, 2017 #7
    isn't machine learning offered only at the grad school level?
     
  9. Feb 21, 2017 #8

    FactChecker

    User Avatar
    Science Advisor
    Gold Member

    You are a beginning programmer. I think that a good approach would be to make sure you are familiar with the computer tools that are readily available for applied statistics. There are free ones like R that you can download. I am sure that there are reasonably priced student licenses for packages like SAS and MATLAB. You will probably have to do this on your own, not as a class.

    Other than that, you can proceed with general programming classes that your school offers as the recommended follow-on for your current introductory class.
     
  10. Feb 22, 2017 #9
    It's offered at the undergrad level at some Universities.
     
  11. Feb 22, 2017 #10
    oh I'm taking "intro to SAS" next semester as part of my primary major (required). isn't SAS obselete?
     
  12. Feb 22, 2017 #11

    FactChecker

    User Avatar
    Science Advisor
    Gold Member

    Very good. I also recommend that you check out R. It's free, well supported and documented, and fairly popular for statistics. You should at least be aware of it.
    SAS is not obsolete. I used it over 35 years ago and even back then, it was more convenient to use than anything I know of today. I haven't used it since then, but I believe that it is still the leading statistics package. I'm sure that it has a lot more capability now than it did when I used it.
     
    Last edited: Feb 22, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted