R help - Apr 2025 - Generate random vectors (continuous number) with a fixed sum

Hi,

There is a function called RandVec in the package Surrogate which can
generate andom vectors (continuous number) with a fixed sum

The help page of this function states that:

a

The function RandVec generates an n by m matrix x. Each of the m
columns contain n random values lying in the interval [a,b]. The
argument a specifies the lower limit of the interval. Default 0.

b

The argument b specifies the upper limit of the interval. Default 1.

However in my case, the lower and upper limits are not same. For
example, if I need to draw a pair of number x, y, such that x + y = 1,
then the lower and upper limits are different.

I tried with below code

library(Surrogate)

RandVec(a=c(0.1, 0.2), b=c(0.2, 0.8), s=1, n=2, m=5)$RandVecOutput

This generates error with message

Error in if (b - a == 0) { : the condition has length > 1

Is there any way to generate the numbers with different lower and upper limits?

RandVec expects both a and b to be single numbers, not numeric vectors
of length > 1.
You could create pairs of (a,b) values and cycle through them. There
are many ways
to do that. Here is one example. (Note that using a=0.1, b=.2, s=1,
n=2 is impossible to achieve, since two numbers bounded above by 0.2
cannot add to 1)

z <- list(c(.1,.2), (.2,.8))
sapply(1:length(z), \(i) RandVec(z[[i]][1], z[[i]][2], s=1, n=2, m=5))

HTH,
Eric


On Mon, Apr 21, 2025 at 5:28?PM Brian Smith <briansmith199312 at
gmail.com> wrote:
>
> Hi,
>
> There is a function called RandVec in the package Surrogate which can
> generate andom vectors (continuous number) with a fixed sum
>
> The help page of this function states that:
>
> a
>
> The function RandVec generates an n by m matrix x. Each of the m
> columns contain n random values lying in the interval [a,b]. The
> argument a specifies the lower limit of the interval. Default 0.
>
> b
>
> The argument b specifies the upper limit of the interval. Default 1.
>
> However in my case, the lower and upper limits are not same. For
> example, if I need to draw a pair of number x, y, such that x + y = 1,
> then the lower and upper limits are different.
>
> I tried with below code
>
> library(Surrogate)
>
> RandVec(a=c(0.1, 0.2), b=c(0.2, 0.8), s=1, n=2, m=5)$RandVecOutput
>
> This generates error with message
>
> Error in if (b - a == 0) { : the condition has length > 1
>
> Is there any way to generate the numbers with different lower and upper
limits?
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

?s 15:27 de 21/04/2025, Brian Smith escreveu:
> Hi,
> 
> There is a function called RandVec in the package Surrogate which can
> generate andom vectors (continuous number) with a fixed sum
> 
> The help page of this function states that:
> 
> a
> 
> The function RandVec generates an n by m matrix x. Each of the m
> columns contain n random values lying in the interval [a,b]. The
> argument a specifies the lower limit of the interval. Default 0.
> 
> b
> 
> The argument b specifies the upper limit of the interval. Default 1.
> 
> However in my case, the lower and upper limits are not same. For
> example, if I need to draw a pair of number x, y, such that x + y = 1,
> then the lower and upper limits are different.
> 
> I tried with below code
> 
> library(Surrogate)
> 
> RandVec(a=c(0.1, 0.2), b=c(0.2, 0.8), s=1, n=2, m=5)$RandVecOutput
> 
> This generates error with message
> 
> Error in if (b - a == 0) { : the condition has length > 1
> 
> Is there any way to generate the numbers with different lower and upper
limits?
> 
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Hello,

Use ?mapply to cycle through all values of a and b.
Note that the output matrices are transposed, the random vectors are the 
rows.


library(Surrogate)

a <- c(0.1, 0.2)
b <- c(0.2, 0.8)
mapply(\(a, b, s, n, m) RandVec(a, b, s, n, m),
        MoreArgs = list(s = 1, n = 2, m = 5), a, b)
#> $RandVecOutput
#>           [,1]      [,2]      [,3]      [,4]      [,5]
#> [1,] 0.2004363 0.1552328 0.2391742 0.1744857 0.1949236
#> [2,] 0.1995637 0.2447672 0.1608258 0.2255143 0.2050764
#>
#> $RandVecOutput
#>           [,1]      [,2]      [,3]      [,4]      [,5]
#> [1,] 0.2157416 0.4691191 0.5067447 0.7749258 0.7728955
#> [2,] 0.7842584 0.5308809 0.4932553 0.2250742 0.2271045


Hope this helps,

Rui Barradas


-- 
Este e-mail foi analisado pelo software antiv?rus AVG para verificar a presen?a
de v?rus.
www.avg.com