[R] Create a vector from another vector

[Duncan Murdoch]

On 8/30/2006 9:49 AM, Doran, Harold wrote:
> Dear list
> 
> Suppose I have the following vector:
> 
> x <-  c(3,4,2,5,6)
> 
> Obviously, this sums to 20. Now, I want to have a second vector, call it
> x2, that sums to x where 5 <= x <= 20, but there are constraints.
> 
> 1) The new vector must be same length as x
> 2) No element of the new vector can be 0
> 3) Element x2[i] of the new vector cannot be larger than element x[i] of
> the original vector
> 4) Ordering is not important
> 
> The following would be an example of what I would want if the user
> wanted the vector x2 to sum to 19
> 
> x2 <- c(2,4,2,5,6) 
> 
> Or, because ordering is not important, this is acceptable
> 
> x2 <- c(3,3,2,5,6) 
> 
> Whereas this would not be appropriate
> 
> x3 <- c(4, 2,2,5,6)
> 
> Because element x3[1] is larger than x[1] even though it sums to 19.

I don't think it's really clear what you mean by "ordering is not 
important".  Would

x2 <- c(6,5,2,4,2)

be acceptable (a re-ordering of your first two examples), even though 
x2[1] > x1[1]?

It's also not clear what result you want in the usual case where there 
are multiple possible answers.  Do you want a randomly chosen one (from 
what distribution?), or would any value satisfying the constraints be 
okay?  (The latter would be easiest:  just start with x1, and decrement 
entries until the desired sum is achieved.  Whether a random value is 
easy or not really depends on the desired distribution.)  Do entries 
need to be integer valued?

> 
> Ideally, the function would take as input the original vector, x, and
> the number that the new vector would sum to. In this example, the vector
> could sum to any number 5 through 20.
> 
> For example,
> 
> myFun <- function(x, sumto) ... details ...
> 
> Is there a preexisiting function that would already do this? I have
> spent too much (unsuccessful) time trying to write a function of my own
> but can't seem to get this to work properly.

I doubt it, because this doesn't look like a standard problem.

Duncan Murdoch

> 
> Any hints would be greatly appreciated.
> 
> Harold
> 
> 
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> 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.