# R: st: constrained sum of random values

 From "Carlo Lazzaro" To Subject R: st: constrained sum of random values Date Thu, 26 Mar 2009 07:52:12 +0100

```Dear Eva and Nick,

Kind Regards,
Carlo

-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Nick Cox
Inviato: mercoledì 25 marzo 2009 20.16
A: statalist@hsphsun2.harvard.edu
Oggetto: RE: st: constrained sum of random values

Carlo specified Stata 9.2; so he would need to use

125 + 25 * invnormal(uniform())

Not rnormal(125,25).

Otherwise Eva's reply shows that Carlo's question is open to various
interpretations

Nick
n.j.cox@durham.ac.uk

Eva Poen

If you put such a constraint on your random values, it means that at
most 76 of them can be random; you need one value to bring the sum up
to 9900. This is very simple to do:

clear
set obs 77
set seed 123
gen n = rnormal(125,25) in 1/76
qui sum n
replace n = 9900-r(sum) in 77

You didn't mention what kind of random values you want, i.e. which
distribution. I assumed normal distribution with mean 125 and sd of
25.

2009/3/25 Carlo Lazzaro <carlo.lazzaro@tiscalinet.it>:

> is there any way for imposing Stata 9.2/SE to create a variable
composed of
> 77 random values so that the sum of these random values is 9900?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```