R help - Aug 2007 - A combinatorial task. How to get rid of loops

Dear List,

I am looking for a faster way of accomplishing this:

# n is an integer larger than two
n <- 3

# matrix of 2^n binary outcomes
bmat <- as.matrix(expand.grid( rep( list(1:0), n))[, n:1])

# I would like to know which rows of "bmat" have "1" in the
"i" and "j"
coordinates, so for n=3 in coordinates 1-2, 1- 3, and 2-3
# I would like then to construct "choose( n, 2)" binary vectors of
lengths
2^n to indicate those rows with a "1"
# The loop below accomplishes this task. Is there a faster way of doing 
the same?

        library(combinat)

        # matrix of all pairwise combinations 
        cmat<-combn( n, 2)
 
        temp<-matrix(NA, nrow(bmat), ncol(cmat))

        for (i in 1:nrow(bmat))
        {
                for (j in 1:ncol(cmat))
                {
                        temp[ i, j]<-as.numeric( bmat[i,][cmat[1,j]] & 
bmat[i,][cmat[2,j]] == 1)
                }
        }


Thank you very much for your help!
Serguei Kaniovski

Serguei Kaniovski wrote:
> Dear List,
> 
> I am looking for a faster way of accomplishing this:
> 
> # n is an integer larger than two
> n <- 3
> 
> # matrix of 2^n binary outcomes
> bmat <- as.matrix(expand.grid( rep( list(1:0), n))[, n:1])
> 
> # I would like to know which rows of "bmat" have "1" in
the "i" and "j"
> coordinates, so for n=3 in coordinates 1-2, 1- 3, and 2-3
> # I would like then to construct "choose( n, 2)" binary vectors
of lengths
> 2^n to indicate those rows with a "1"
> # The loop below accomplishes this task. Is there a faster way of doing 
> the same?
> 
>         library(combinat)
> 
>         # matrix of all pairwise combinations 
>         cmat<-combn( n, 2)

One example is:

n <- 3
bmat <- as.matrix(expand.grid( rep( list(1:0), n))[, n:1])
library(combinat)
cmat<-combn( n, 2)
apply(cmat, 2, function(x) as.numeric(bmat[,x[1]] & bmat[,x[2]]))

Uwe Ligges


>         temp<-matrix(NA, nrow(bmat), ncol(cmat))
> 
>         for (i in 1:nrow(bmat))
>         {
>                 for (j in 1:ncol(cmat))
>                 {
>                         temp[ i, j]<-as.numeric( bmat[i,][cmat[1,j]]
&
> bmat[i,][cmat[2,j]] == 1)
>                 }
>         }
> 
> 
> Thank you very much for your help!
> Serguei Kaniovski
> 
> ______________________________________________
> 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
> and provide commented, minimal, self-contained, reproducible code.