Re: st: Drawing multiple samples from a normal distribution

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 11:12 AM 12/21/2004 -0500, Joseph Cordes wrote:
> Actually, because we don't want them to actually know the "true" means and 
> SDs we are going to generate the samples ourselves.  Do I understand you 
> correctly to say that each time we make a draw we need to provide a 
> different seed?  And if the answer is yes, am I correct in assuming that 
> the seed can be any number chosen?
Here is another variation that might meet your needs, depending on exactly 
what you want:

forval i = 1/25 {
   clear
   set obs 100
   set seed `i'
   drawnorm sample`i', m(25) sd(5)
   save sample`i'
}

That will give you 25 different samples.  I like my earlier variation where 
everybody has access to all the data but maybe you don't want that.

With setting seeds, you want to be careful that (a) you don't draw the 
exact same samples when you don't want to.  For example, if you just gave 
people commands to draw the sample themselves, then if they all gave those 
exact same commands immediately after starting Stata, then I believe they 
would all generate the same sample.  (b) However, once people start varying 
in their commands, then the seeds start changing around, and then you get 
the problem that you may not be able to reproduce something because you 
don't know what the seed was.  Setting the seed yourself (using different 
but known values) helps to avoid such problems.



-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/