[R] recommended combo of apps for new user?
jrkrideau at yahoo.ca
Mon Aug 20 16:52:34 CEST 2007
--- hadley wickham <h.wickham at gmail.com> wrote:
> On 8/18/07, John Kane <jrkrideau at yahoo.ca> wrote:
> > I'm just starting to get a grasp on how R works so
> > don't take my words too seriously but have a look
> > http://addictedtor.free.fr/graphiques/ for some
> > of what R can do for publication quality graphics.
> > is always possible that you might need another
> > graphics package as well but I think it unlikely.
> I may be in the minority, but I really don't like
> the R graph gallery.
> To my eye it largely provides examples of what you
> _shouldn't_ do
> with graphics (and also seems rather unloved at the
> moment, given the
> large number of spam keywords). It fails to provide
> examples of using
> graphics to gain insight into your data and mainly
> focuses on drawing
> pretty (ugly) pictures.
> Unfortunately there aren't many better resources at
> the moment.
> Deepayan Sarkar is working on a lattice book, and
> hopefully he will
> make the plots available on his website as well.
> I'm also working on
> a book for my ggplot2 package
> (http://had.co.nz/ggplot2) but that
> won't be finished until next year. For interactive
> graphics, the
> GGobi book (http://www.ggobi.org/book/) is very
> close to being
> published, and provides details about the R-GGobi
> link as well as many
> techniques for gaining insight into your data
> interactively. Another
> option is the Graphics of Large Dataset book
> which provides a wider survey of state of the art in
> graphics for large datasets.
I don't disagree in a number of cases but it is a good
example of the myriad of things that one can do with R
graphics even if you may not want to! And in some
cases, expecially in the business world, jazzing up a
graphic may be needed even if it does not improve the
actual understanding of the data.
In the context of the OP's question about graphical
capabilites for publication quality output I think the
gallery is not a bad illustration of R's versitility.
Thanks for the book references. Now just to pry them
loose from wthe library; well the Cook & Swayze one
isn't through cataloging yet.
More information about the R-help