[R] [Johannes Huesing <tmi0m0@sp2.power.uni-essen.de>] Media coverage

Ulf Mehlig umehlig at uni-bremen.de
Mon Nov 30 09:31:17 CET 1998

Douglas Bates <bates at stat.wisc.edu> wrote:

> Can anyone summarize what was said about R in the article?
[from Linux Magazin]

The article is a survey of software for statistics ("Statistical data
analysis with Linux"). The author (Dr. Wenzel Matiaske,
<wm at perform.ww.tu-berlin.de>, address according to Linux Magazin)
gives informations about

      S-Plus 5
      ViSta (and XLisp-Stat)
      R (0.62.2)

I haven't time to translate everything (and that's better for the
native English speaker, anyway ;-) ...


About R he says (summarised!!): R is like XploRe a statistical
programming language and system for graphical and numerical data
analysis. Was written by ... similar to S ... can use code written for
S and S' documentation ... existence of CRAN ... availability as
deb/rpm package, additional libraries are on CRAN, too ... some
installation hints (FORTRAN translator, Perl5 necessary). R has a
simple interactive user interface ... demo ()/help () ... more
comfortable emacs interface exists ... graphical representation
possible with Xgobi (screen-shot with emacs & Xgobi). As example of R
there is a box with code showing how to make a histogram, a
correlation analysis and a linear regression; the code is explained in
the text. Number of existing procedures is considered impressive
... base package contains all possibilities of univariate analysis and
many methods for statistical modelling (linear Model, generalised
linear model) ... in the libraries there are: ctests, tree, cluster,
eda, mva, multiv, MASS (purpose of the modules is explained
shortly). Something is still missing: "Kreuztabellenanalyse" -- cross
table analysis? --, MANOVA, factor analysis, general additive
models. But the number of functions is increasing, and there are
Fortran/C interfaces. R not tailored to data storing, importing,
report generation, in this respect needs help of a database or another
data analysis system.

The author summarises: XploRe and R are recommended as programming
environments and contain the biggest number of statistical models
(there is a table comparing all programs by supported statistical
methods and user interface facilities).


Please excuse for translation errors and misunderstandings (I'm no
expert in statistics!).

Hope it helps ...

 %%%%%            Ulf Mehlig              <umehlig at zmt.uni-bremen.de>
   %%%%!%%%       Projekt "MADAM"         <umehlig at uni-bremen.de>
%%%% %!% %%%%     ----------------------------------------------------
 ---| %%%         MADAM:  MAngrove    |  Center for Tropical Marine
    ||--%!%              Dynamics     |  Biology
    ||                  And           |  Fahrenheitstrasse 1
 _ /||\_/\_            Management     |  
/  /    \  \ ~~~~~~~~~~~~~~~~~        |  28359 Bremen/Germany
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list