Here are the functions documented in the Blue Book that I found missing
in R (ignoring the ones which are obviously outdated).
aggregate allocated amatch axes chull clorder cutree cycle date
debugger dget discr faces interp l1fit labclust lag loglin
monthplot mstree mulbar napsack odometer persp plclust plotfit
rep.int restore rreg sabl sablplot set.seed smooth sort.list
Stable stars starsymb storage subtree symbols tempfile tslines
tsmatrix tsplot tspoints twoway unlink usa xor
In some detail:
aggregate cycle lag smooth tslines tsmatrix tsplot tspoints:
Obviously useful for time series. Andreas has a quick version
of lag() ...
allocated storage:
Something to display the current memory consumption (other than
as a byproduct of calling gc() would be nice ...
amatch
axes:
Seems to be the same as title() except for axes = T and F, resp.
R currently does not have this extra arg.
chull
clorder(?) cutree discr faces(?) labclust(?) mstree(?) plclust stars
subtree:
These are already (sort of) in the full version of `mva' as
(still?) planned by Ross (question marks indicate the ones not
mentioned explicitly)
date:
If we'd want this, we could add
date <- function() system("date")
to work under UNIX, at least
debugger
dget
loglin:
Would this be useful? It should not be too hard to write
something around AS 51 ... Should that be compatible with the S
version?
interp
l1fit rreg:
These would certainly be nice as workhorses, but shouldn't L1
and robust fitting of linear models be specifyable at the model
level?
monthplot sabl sablplot
mulbar:
Would be nice to have, in particular in connection with Mosaic
plots ...
napsack
odometer
persp:
This is already in "3D GRAPHICS" of PROJECTS.
plotfit twoway:
Outdated?
rep.int
restore:
Should be the opposite of dump() ... In R, can we simply use
source()?
set.seed:
Martin, do you have a version of that for R?
sort.list
Stable:
Jim Lindsey <jlindsey@luc.ac.be> wrote that he has d and p
function for the four-parameter stable family, perhaps we should
include this.
starsymb
symbols
tempfile:
Would be very useful, although I'd prefer it to be called
`tmpfile' instead. I am not sure if tmpfile() is ANSI C. If
so, we could simply wrap around it, as Fritz posted some time
ago.
unlink:
We most likely don't need this one, right?
usa:
Would be nice, of course ...
xor:
Is trivial and useful. Could we add
xor <- function(x, y) { (x | y) & !(x & y) }
to the distribution?
-k
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel 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-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
