[R] What's the BEST way in R to adapt this vector?

Gabor Grothendieck ggrothendieck at gmail.com
Sun Nov 23 05:43:30 CET 2008


Sorry, you wanted a vector not a matrix.
Here are several possibilities. (If you are
willing to hard code the levels then they
can be shortened by replacing sort(unique(y))
or levels(factor(y)) with its value -- in this
case 1:3)

c(outer(sort(unique(y)), y, "=="))+0

c(outer(levels(factor(y)), y, "=="))+0

levs <- sort(unique(y))
rep(y, each = length(levs)) == levs

c(sapply(y, "==", sort(unique(y)))) + 0

c(rbind(y == 1, y == 2, y ==3))+0

On Sat, Nov 22, 2008 at 10:12 PM, Gabor Grothendieck
<ggrothendieck at gmail.com> wrote:
> Try this:
>
> outer(y, sort(unique(y)), "==")+0
>
>
> On Sat, Nov 22, 2008 at 3:37 PM, zerfetzen <zerfetzen at yahoo.com> wrote:
>>
>> Goal:
>> Suppose you have a vector that is a discrete variable with values ranging
>> from 1 to 3, and length of 10.  We'll use this as the example:
>>
>> y <- c(1,2,3,1,2,3,1,2,3,1)
>>
>> ...and suppose you want your new vector (y.new) to be equal in length to the
>> possible discrete values (3) times the length (10), and formatted in such a
>> way that if y[1] == 1, then y.new[1:3] == c(1,0,0), and if y[2] == 2, then
>> y.new[4:6] == c(0,1,0).  For example, the final goal should be:
>>
>> y.new <- c(1,0,0,0,1,0,0,0,1,1,0,0,0,1,0,0,0,1,1,0,0,0,1,0,0,0,1,1,0,0)
>>
>> Note: I know how to do this with loops, but that's not taking advantage of
>> R's capabilities with vectors and, I suspect, matrices.
>>
>> So far, here's my best:
>>
>> y <- c(1,2,3,1,2,3,1,2,3,1)
>> y.k <- length(unique(y)) #Num. of Categories of y
>> y.n <- NROW(y)
>> y.nk <- y.n * y.k #Length of new vector
>> y.1 <- ifelse(y == 1,1,0)
>> y.2 <- ifelse(y == 2,1,0)
>> y.3 <- ifelse(y == 3,1,0)
>> z <- cbind(y.1, y.2, y.3)
>> z.trans <- t(z)
>> y.new <- c(1:y.nk); y.new[1:y.nk] <- NA
>> y.new <- as.vector(z.trans)
>> rm(y.k, y.n, y.nk, y.1, y.2, y.3, z, z.trans)
>> y
>> y.new
>>
>> Is there a better a way?  I'm still pretty new.  Thanks.
>> --
>> View this message in context: http://www.nabble.com/What%27s-the-BEST-way-in-R-to-adapt-this-vector--tp20638991p20638991.html
>> Sent from the R help mailing list archive at Nabble.com.
>>
>> ______________________________________________
>> 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.
>>
>



More information about the R-help mailing list