st: Prediction at the 'mean' of a categorical- xi3?

[graupel75@gmail.com]

Dear Statalisters-

I am attempting to do predictions from a model which includes clinical
isolates as a predictor; these are important for good model fit, but
not useful in and of themselves. With a continuous you would likely
just set the reference to the median/mean, and predict away. Any
suggestions in how to code it so I get something similar (a population
average value) for a categorical in the presence of other covariates?

Michael Mitchell & Phil Ender's -xi3- (ssc) has lots of flexibility
and options, but I am unsure of how to implement...

thank you-

Andrew Lover
______________________________________________
Epidemiologist
Centre for Infectious Disease Epidemiology Research (CIDER)
Saw Swee Hock School of Public Health
National University of Singapore
*
*   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/