Re: st: estimating data for a given correlation coefficient

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

At 09:54 AM 1/20/2005 -0500, Nomoreno wrote:
> First you need to define the correlation matrix you desire, then you can 
> use -drawnorm- in STATA to generate the data you wish for different panels 
> of the variable, given the correlation matrix. For more information, you 
> can check out -drawnorm- in STATA.
>
> I hope it helps.
Perhaps better is -corr2data-.  -drawnorm- will draw a random sample from a 
population with the correlations you give it; because of sampling 
variability the sample correlations will not be identical to the population 
correlations.  If that is what you want, then fine.  But, if you want a 
data set that has exactly the correlations you specify, then use 
-corr2data-.  This might be the case if, say, you want to replicate the 
results from a published analysis where the correlations are included in 
the text.

A discussion of the use of -corr2data- can be found on pp. 8-10 of

http://www.nd.edu/~rwilliam/stats1/OLS-Stata.pdf


-------------------------------------------
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/