Radified Community Forums | |
http://radified.com/cgi-bin/yabb2/YaBB.pl
Rad Community Non-Technical Discussion Boards >> YaBB Forum Software + Rad Web Site >> R Programming language for statistical computing http://radified.com/cgi-bin/yabb2/YaBB.pl?num=1214528040 Message started by Rad on Jun 26^{th}, 2008 at 7:54pm |
Title: R Programming language for statistical computing Post by Rad on Jun 26^{th}, 2008 at 7:54pm
Speaking of programming languages .. have you heard of 'R'? (which is an open-source version, based on 'S')
http://www.r-project.org/ http://en.wikipedia.org/wiki/R_(programming_language) .. which is supposedly used for "statistical computing"? I'm guessing the 'S' stands for statistical. what types of things would this language be used for? |
Title: Re: R Programming language for statistical computing Post by Nigel Bree on Jun 26^{th}, 2008 at 11:25pm
Overall, it'd be used for similar things to SPSS (Statistical Package for Social Sciences) and SAS, which are pieces of statistical analysis software that have been around since the mainframe days. Wrangling data, looking to verify hypothesis and generating graphs to help explain your results.
Most of these things include some form of general programming facility, since it's the nature of things that you want to re-run particular analyses on different data sets and compare results, and statistical tools generally combine these with some particular output and visualization facilities to compare the transformed data sets. Being specialized for statistics is reasonable since it's a sub-field of mathematics which has some unique aspects compared to the kind of thing you use in tools for general engineering and science (where of course, you have Mathematica, which includes its own programming language that has grown over the years but was originally modeled on Lisp concepts) . Time-series work in particular is a whole sub-field; forecasting is a particular application of time-series analysis that is worthy of entire textbooks (that one in particular I've read and quite enjoyed, in the second edition). |
Title: Re: R Programming language for statistical computing Post by zmdmw52 on Jul 3^{rd}, 2008 at 11:26pm Rad wrote on Jun 26^{th}, 2008 at 7:54pm:
This ppt compares SPSS vs SAS vs 'R' (see slides 14 thro' 19). SAS has a somewhat steep learning curve, a background in statistics &/or programming (particularly the latter) helps. Though I haven't used 'R' (and am just about familiar with very basic-level SAS syntax towards data analysis). From what I've heard from those using 'R', it is still more difficult, in view of less help/ documentation available. Also, not sure if 'R' has as many modules as does SAS; which also has extensive online documentation and numerous books available for various levels. The Little SAS Book gives a good overview, this book published by SAS Publishing is also a good introduction. |
Radified Community Forums » Powered by YaBB 2.4! YaBB © 2000-2009. All Rights Reserved. |