[Rd] text.rpart: Unwanted NA labels on terminal nodes (PR#1009)

[Prof Brian D Ripley]

On Tue, 3 Jul 2001 john.maindonald@anu.edu.au wrote:

> Brian
> The following (which is new to rw1030) occurs with both
> Windows 98 & Windows ME.  I have not tested behaviour
> under Unix or Linux, but I expect it is no different.

I think you mean this is a change from rpart 2.0-2 to 3.0-0: packages have
version numbers too!

It's a R/S difference (I would say an R bug) for which the workaround got
lost.  At the top of text.rpart insert

    FUN1 <- function(x, y, lab,...){
        ind <- lab != "NA"
        FUN(x[ind], y[ind], lab[ind], ...)
    }

and replace FUN by FUN1 inside if(splits) {} (but not at the end).

Fixed in rpart 3.0-1, so update.packages() will give this to you shortly.

The difference can be seen in

> plot(1:3)
> text(2, 3, NA)

S's behaviour is much more convenient and logical.


> text.rpart() prints unwanted NAs (presumably in the
> splitting criterion position) on terminal nodes.
>
>           Criterion <- factor(paste("Leaf", 1:5))
>           Node <- factor(1:5)
>           assign("tree.df", data.frame(Criterion = Criterion, Node = Node))
>           nobs <- dim(tree.df)[[1]]
>           u.tree <- rpart(Node ~ Criterion, data = tree.df, all = F,
> control = list(minsplit = 2, minbucket
>                    = 1, cp = 9.999999999999998e-008))
>           plot(u.tree, uniform=T)
>           text(u.tree)
>
>
> --please do not edit the information below--
>
> Version:
>   platform = i386-pc-mingw32
>   arch = x86
>   os = Win32
>   system = x86, Win32
>   status =
>   major = 1
>   minor = 3.0
>   year = 2001
>   month = 06
>   day = 22
>   language = R
>
> Windows 98 SE 4.10 (build 2222)  A
>
> Search Path:
>   .GlobalEnv, package:rpart, package:ctest, Autoloads, package:base
> John Maindonald               email : john.maindonald@anu.edu.au
> Statistical Consulting Unit,  phone : (6125)3998
> c/o CMA, SMS,                 fax   : (6125)5549
> John Dedman Mathematical Sciences Building
> Australian National University
> Canberra ACT 0200
> Australia
>
>
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> 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
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
>

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._