# Re: [R] sampling vectors

From: Daniel Nordlund <res90sx5_at_verizon.net>
Date: Wed 05 Oct 2005 - 02:02:58 EST

> -----Original Message-----
> From: r-help-bounces@stat.math.ethz.ch [mailto:r-help-bounces@stat.math.ethz.ch]
> On Behalf Of ecatchpole
> Sent: Monday, October 03, 2005 11:57 PM
> To: Eric Pante
> Cc: r-help@stat.math.ethz.ch
> Subject: Re: [R] sampling vectors
>
> Eric,
>
> If you want samples of size 3 from 0:50, with sum==50, this seems to do
> the job (with apologies to those who really know how to program in R):
>
> tot <- 50
> ii <- 0
> aa <- list()
> for(i in 0:tot){
> for(j in 0:(tot-i)){
> k <- tot-i-j
> ii <- ii+1
> aa[[ii]] <- list(i=i,j=j,k=k)
> }
> }
>
> aa[sample(ii, 4)] # for a sample of 4.
>
> If you want a sample of size n from x[1],...x[N], such that the sample
> sum is T, then that is much trickier!
>
> Ted.
>
> On 04/10/05 07:31, Eric Pante wrote,:
> > Hello Listers,
> >
> > I am trying to sample a vector to create a new one of sample length,
> > witha sum equal to the sum of the initial vector:
> >
> > initial = 10, 30, 10 (sum=50)
> > sample example = 5, 35, 10 (sum=50) or 25, 15, 10 (sum=50), etc ...
> >
> > My problem is to control the sum, so it stays constant.
> > Any suggestions would be very helpful !
> >
> > Thank you in advance,
> > Eric
> >

Eric,

Here is another solution, which allows vectors of different lengths and sums.

vectorSample <- function(vec) {

```  tot<-sum(vec)
Len<-length(vec)
v < -rep(0,Len)
```

for(i in Len:2) {
UL <- tot - sum(v) - i + 1
v[i]<-sample(1:UL,1)
}
v[1] <- tot - sum(v)
v
}

vectorSample( c(10,30,10) )

Hope this helps,

Dan Nordlund
Bothell, WA

R-help@stat.math.ethz.ch mailing list