[R] Loop removal?

[Peter Malewski]

I don't know if I understand you right, but..


> xx <- function(x,g=2)rep(1:g,rep(length(x) / g  , g ) )
> zz <- data.frame (data=rnorm(100), id=rep(1:10,10))
> cbind(zz,nn=unlist( tapply(zz$data,zz$id,function(x)xx(x[order(x)] ,
g=2) )))[1:13,]
          data id nn
1  -1.06975967  1  1
2   1.30402032  2  1
3  -0.08553785  3  1
4   0.11884050  4  1
5  -0.10690368  5  1
6   1.13232754  6  2
7   1.97065424  7  2
8  -0.19009641  8  2
9  -0.72320338  9  2
10 -0.53456589 10  2
11 -0.01734236  1  1
12  0.04074360  2  1
13  2.28373113  3  1
> 
 

BUT NOTE, that this only works if 

length(x) %% g == 0

...
Peter

On Sun, 30 Jul 2000 rijn at swi.psy.uva.nl wrote:

> Hi,
> 
> > Look at the help for the cut() function, which does what you want, 
> > I think.
> 
> Thanks for the suggestion. I'm sure I can put it to good use in another
> part of the analysis, but it does not exactly what I want for this problem
> (as far as I understand). Given the following example:
> 
> x <- c(1,1,2,3,4,4,3,2,1,1)
> 
> And assuming I want to divide x into to equally long sets to which the
> values get assigned in increasing order, cut doesn't do the job:
> 
> > cut(x,c(1,2,4),labels=F,include.lowest=T)
>  [1] 1 1 1 2 2 2 2 1 1 1
> > table(cut(x,c(1,2,4),labels=F,include.lowest=T))
> 
> 1 2 
> 6 4 
> > 
> 
> Whereas I'm looking for an output that results in:
> 
> > table(...(x,...))
> 
> 1 2
> 5 5
> >
> 
> So, the assignment of factors is not only based on the size of the numbers
> in the vector, but also on the lengths of the resulting vectors.
> 
> And, it still leaves this "loop"-problem:
> 
> > > for (i in unique(stor$subject)) {
> > >   stor[stor$subject == i,]$bin <- bindata(stor[stor$subject == i,]$rt)
> > > }
> 
> I assume I should use a function like tapply, but:
> 
> > bindata2 <- function(rt) {
> +   
> +   bindata <- rep(-1,length(rt))
> +   binwidth <- length(rt)/2
> +   
> +   bindata[order(rt)[(0*binwidth)+1:(1*binwidth)]] <- 1
> +   bindata[order(rt)[(1*binwidth)+1:(2*binwidth)]] <- 2
> + 
> +   bindata
> + }
> > x <- data.frame(subject=c(1,1,1,2,2,2),rt=c(1,2,3,3,2,4))
> > tapply(x$rt,x$subject,bindata2)
> $"1"
> [1] 1 2 2
> 
> $"2"
> [1] 2 1 2
> 
> the problem is that I don't know how to combine/concatenate the output of
> tapply into one vector again. So, it's probably all caused by a very
> simple "lack of knowledge about R" on my side...
> 
> Right now, I use tapply, but use a for-loop again to concatenate the list
> into a single vector again. But it feels like there should be a better
> solution. I someone can hint me towards one, I would be very pleased.
> 
>    Hedderik.
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help 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-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> 

**************************************************************************
P.Malewski				Tel.: 0531 500965
Maschplatz 8				mailto: Peter.Malewski at gmx.de
************************38114 Braunschweig********************************

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._