[R] binary order combinations

[Henrik Bengtsson]

names <- sprintf("V%d", 1:4);
n <- length(names);
stopifnot(n <= 32); # Theoretical upper limit
x <- matrix(intToBits(1:(2^n-1)), ncol=2^n-1);
x <- x[1:n,,drop=FALSE];
keys <- apply(x, MARGIN=2, FUN=function(z) paste(names[as.logical(z)],
collapse=":"));
print(keys);

 [1] "V1"          "V2"          "V1:V2"       "V3"
 [5] "V1:V3"       "V2:V3"       "V1:V2:V3"    "V4"
 [9] "V1:V4"       "V2:V4"       "V1:V2:V4"    "V3:V4"
[13] "V1:V3:V4"    "V2:V3:V4"    "V1:V2:V3:V4"

/H

On Fri, Sep 5, 2008 at 8:23 AM, Dimitri Liakhovitski <ld7631 at gmail.com> wrote:
> I am not sure it can do it. Besides, I ran a test of combos from quantreg:
>
> library(quantreg)
> H<-1:3
> test.combos<-lapply(1:3,function(x)
> {combn(H,x)
> })
>
> Every time I tried it crashed my R...
>
> :(
>
> Dimitri
>
> On 9/5/08, roger koenker <rkoenker at uiuc.edu> wrote:
>> Does ?combos in the quantreg package do what you want?
>>
>>
>> url:    www.econ.uiuc.edu/~roger            Roger Koenker
>> email    rkoenker at uiuc.edu            Department of Economics
>> vox:     217-333-4558                University of Illinois
>> fax:       217-244-6678                Champaign, IL 61820
>>
>>
>>
>>
>> On Sep 5, 2008, at 9:58 AM, Dimitri Liakhovitski wrote:
>>
>> >
>> > Dear all!
>> >
>> > I have a vector of names
>> > names<-("V1", "V2", "V3",....., "V15")
>> >
>> > I could create all possible combinations of these names (of all
>> > lengths) using R:
>> >
>> > combos<-lapply(1:15,function(x)
>> > {combn(names,x)
>> > })
>> >
>> > I get a list with all possible combinations of elements of 'names'
>> > that looks like this (just the very beginning of it):
>> >
>> > [[1]] - the first element contains all combinations of 1 name
>> >    [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13]
>> [,14]
>> > [1,] "V1" "V2" "V3" "V4" "V5" "V6" "V7" "V8" "V9" "V10" "V11" "V12" "V13"
>> "V14"
>> >    [,15]
>> > [1,] "V15"
>> >
>> > [[2]] - the second element contains all possible combinations of 2 names
>> >    [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]  [,10] [,11] [,12] [,13]
>> > [1,] "V1" "V1" "V1" "V1" "V1" "V1" "V1" "V1" "V1"  "V1"  "V1"  "V1"  "V1"
>> > [2,] "V2" "V3" "V4" "V5" "V6" "V7" "V8" "V9" "V10" "V11" "V12" "V13" "V14"
>> > .
>> > .
>> > .
>> > etc.
>> >
>> > My question is: Is there any way to re-arrange all sub-elements of the
>> > above list (i.e., all possible combinations of names such as V1,
>> > V1:V3, V1:V2:V4:V5) in a binary system order. More specifically,
>> > according to this system:
>> > V1=1
>> > V2=2
>> > V3=4
>> > V4=8
>> > V5=16, etc....
>> >
>> > So, I'd like those combinations to be arranged in a vector in the
>> > following order:
>> > 1. V1 (because V1=1)
>> > 2. V2 (because V2=2)
>> > 3. V1:V2 (because V1=1 and V2=2 so that 1+2=3)
>> > 4. V3 (because V3=4)
>> > 5. V1:V3 (because V1=1 and V3=4 so that 1+4=5)
>> > 6. V2:V3 (because V2=2 and V3=4 so that 2+4=6)
>> > 7. V1:V2:V3 (because V1=1 and V2=2 and V3=4 so that 1+2+4=7)
>> > 8. V4 (because V4=8)
>> > etc.
>> >
>> > Is it at all possible?
>> > Or maybe there is a way to create the name combinations in such an
>> > order in the first place?
>> >
>> > Thank you very much!
>> > Dimitri Liakhovitski
>> >
>> > ______________________________________________
>> > R-help at r-project.org 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.
>> >
>>
>>
>
> ______________________________________________
> R-help at r-project.org 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.
>