Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: constrained sum of random values


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: constrained sum of random values
Date   Wed, 25 Mar 2009 19:15:44 -0000

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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index