Re: st: Confidence Intervals of Predicted scores in CLARIFY - adding together

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Fri, 3/9/10, James Laurence wrote:
> I want to create a graph in excel using the top two
> categories combined: so 0.674 + 0.183 = this gives me a
> predicted score of 0.86
> 
> Here's  my question. I also want to add CONFIDENCE
> INTERVALS to this value using the C.I.s provided by CLARIFY.
> Do I simply add the C.I.'s together for the top two
> categories as I did with the predicted scores? 
> e.g. the C.I.s for this predicted score would be min. 0.656
> + 0.169  and the max. would be 0.691 + 0.199?
> 
> So, the C.I.s of my predicted score of 0.86 (minimum 
> 0.825) and (maximum 0.888)? 

No, that way you are ignoring the correlations in the sampling
distribution. I don't think -clarify- returns a full variance
covariance matrix (though I could be wrong). In that case
the easiest solution is probably to run your simulations again,
but this time explicitly sampling the parameters that interest
you.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/