I often use the following extension of Christian's suggestion:

> nam <- paste("var", 1:5, sep="")
> nam
[1] "var1" "var2" "var3" "var4" "var5"
> nam. <- paste(nam, collapse="+")
> nam.
[1] "var1+var2+var3+var4+var5"
> nam2 <- paste("I(", nam, "^2)", sep="")
> nam2
[1] "I(var1^2)" "I(var2^2)" "I(var3^2)" "I(var4^2)" "I(var5^2)"
> nam2. <- paste(nam2, collapse="+")
> nam2.
[1] "I(var1^2)+I(var2^2)+I(var3^2)+I(var4^2)+I(var5^2)"
> mdl2 <- paste("y", "~ (", nam., ")^2 +", nam2.)
> mdl2
[1] "y ~ ( var1+var2+var3+var4+var5 )^2 + I(var1^2)+I(var2^2)+I(var3^2)+I(var4^2)+I(var5^2)"
> formula(mdl2)
y ~ (var1 + var2 + var3 + var4 + var5)^2 + I(var1^2) + I(var2^2) +
I(var3^2) + I(var4^2) + I(var5^2)


This will fit a full quadratic in the 5 variables.

If Uwe's solution won't work for you, this can be modified to work with any vector of appropriate variable names. hope this helps. spencer graves

[EMAIL PROTECTED] wrote:

On 16 Apr 2004 at 17:08, Christian Hennig wrote:

You can build your formula along the ways of the following example, without typing the 200 names:



nam <- paste("var", 1:5, sep="", collapse="+")
nam


[1] "var1+var2+var3+var4+var5"


nam <- paste("y", nam, sep="~")
nam


[1] "y~var1+var2+var3+var4+var5"


as.formula(nam)


y ~ var1 + var2 + var3 + var4 + var5

Kjetil Halvorsen



Hi,

is it true that the only way to tell rpart which variables to use is
to specify every single variable in the formula and separate them with
"+"?

I wonder if this error...



rotree <- rpart(powerdata[,1]~powerdata[,2:3],


+ method="class")
Error in model.frame(formula, rownames, variables, varnames, extras,
extranames, : invalid variable type


...can be prevented by other means than...



rotree <- rpart(segments~V1.3+V1.9,


+ method="class",data=powerdata)

This works.
(The variables here are powerdata[,1] and [,2:3].)

But I do not want to write 200 variable names down, as this is the
real size of "powerdata".

Is there any way around that?

Christian



**********************************************************************
* Christian Hennig Fachbereich Mathematik-SPST/ZMS, Universitaet
Hamburg [EMAIL PROTECTED],
http://www.math.uni-hamburg.de/home/hennig/
######################################################################
# ich empfehle www.boag-online.de

______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html



______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html



______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

Reply via email to