Re: st: Using/Accessing discriminant function scores?

[Richard Goldstein <richgold@ix.netcom.com>]

I'm not clear exactly what you want but try this

-h candisc-
click on "also see" and then on "discrim lda postestimation"
look for the "predict" option and click on it to see what is available

Rich

On 10/3/13 1:17 PM, Wilson, Kimberly J wrote:
> I am running some discriminant function analyses in STATA.  I've run the core analyses using the candisc function. 
> 
> I've got more than 2 groups so need to test which groups significantly differ from one another on the function means (group centroids) using oneway ANOVA and either the Scheffe or Bonferroni option.
> In order to do that, however, I need to know how individual respondents score on each of the discriminant functions.
> The candisc function provides a table of the by-group means on each of the functions, so I know that the information I need must be saved somewhere, but cannot figure it out from the STATA manual or help files.
> 
> Thus far, I've been calculating the scores manually from the standardized canonical DF coefficients.  But we've got 12 potentially discriminating variables & 2 functions, and I'd like to minimize the chances of mistakes by generating the scores calculated by STATA.  If anyone knows how to do this, I'd appreciate your insights.
> 
> 
> Kim Wilson, MPIA
> 2012-2013 Dell Health Scholar
> Michael & Susan Dell Center for Healthy Living
> University of Texas School of Public Health
> Austin Regional Campus
> 1616 Guadalupe Street, Suite 6.330, Austin, TX 78701
> kimberly.j.wilson@uth.tmc.edu
> cell: 412/ 719.9555
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/