R help - Jun 2005 - quick way to construct formula

Dear R users,

I have a data with 1000 variables named "x1", "x2", ...,
"x1000", and
I want to construct a formula like this format:

~x1+x2+...+x1000+x1:x2+x1:x3+x999:x1000+log(x1)+...+log(x1000)

That is: the base variables followed by all interaction terms and all
base feature log-transformations. I know I can use several paste
functions to construct it. But is there any other handy way to do it?

-Luke

Here is a way: it uses 'paste' but I dont think it is a problem anyway
to use it.
Nevertheless, it is surely a bad idea to fit any model with more than
250000 terms...

>   main_effects = paste(nam,collapse="+")
>   inter <- outer(nam,nam,paste,sep=":")
>   inter <- paste(inter[upper.tri(inter)],collapse="+")
>   log_effects <-
paste("log(",nam,")",sep="",collapse="+")
>
as.formula(paste("~",main_effects,"+",inter,"+",log_effects,sep=""))
~x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x1:x2 + x1:x3 + 
    x2:x3 + x1:x4 + x2:x4 + x3:x4 + x1:x5 + x2:x5 + x3:x5 + x4:x5 + 
    x1:x6 + x2:x6 + x3:x6 + x4:x6 + x5:x6 + x1:x7 + x2:x7 + x3:x7 + 
    x4:x7 + x5:x7 + x6:x7 + x1:x8 + x2:x8 + x3:x8 + x4:x8 + x5:x8 + 
    x6:x8 + x7:x8 + x1:x9 + x2:x9 + x3:x9 + x4:x9 + x5:x9 + x6:x9 + 
    x7:x9 + x8:x9 + x1:x10 + x2:x10 + x3:x10 + x4:x10 + x5:x10 + 
    x6:x10 + x7:x10 + x8:x10 + x9:x10 + log(x1) + log(x2) + log(x3) + 
    log(x4) + log(x5) + log(x6) + log(x7) + log(x8) + log(x9) + 
    log(x10)




HTH,

Eric



Eric Lecoutre
UCL /  Institut de Statistique
Voie du Roman Pays, 20
1348 Louvain-la-Neuve
Belgium

tel: (+32)(0)10473050
lecoutre at stat.ucl.ac.be
http://www.stat.ucl.ac.be/ISpersonnel/lecoutre

If the statistics are boring, then you've got the wrong numbers. -Edward
Tufte   

> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Luke
> Sent: mercredi 29 juin 2005 4:49
> To: R-help at stat.math.ethz.ch
> Subject: [R] quick way to construct formula
> 
> 
> Dear R users,
> 
> I have a data with 1000 variables named "x1", "x2",
...,
> "x1000", and I want to construct a formula like this format:
> 
> ~x1+x2+...+x1000+x1:x2+x1:x3+x999:x1000+log(x1)+...+log(x1000)
> 
> That is: the base variables followed by all interaction terms 
> and all base feature log-transformations. I know I can use 
> several paste functions to construct it. But is there any 
> other handy way to do it?
> 
> -Luke
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
>

On Tue, 28 Jun 2005, Luke wrote:
> Dear R users,
>
> I have a data with 1000 variables named "x1", "x2",
..., "x1000", and
> I want to construct a formula like this format:
>
> ~x1+x2+...+x1000+x1:x2+x1:x3+x999:x1000+log(x1)+...+log(x1000)
>
> That is: the base variables followed by all interaction terms and all
> base feature log-transformations. I know I can use several paste
> functions to construct it. But is there any other handy way to do it?
Do you really want a formula with over half a million terms in?  I think 
it is likely that R will hit recursion limits in handing such a formula, 
quite possibly C-level stack overflow.

For a more modest example:

dd <- data.frame(x1=1, x2=2, x3=3)
terms(~ . + .^2, data=dd, simplify=TRUE)

will get you all terms and their interactions.  Using paste() to get the 
log terms is the easiest way I know.


-- 
Brian D. Ripley,                  ripley at 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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595