st: standardized confidence intervals using mlogit

["Staimez, Lisa Rachel" <lrlevy@emory.edu>]

Dear Listers,
I have been trying  polychotomous logistic regression, modeling an outcome with 5 levels, group, and I am coming across two problems.

First, I am unable to obtain the SE and confidence interval of standardized coefficients. The list coef, std  command does not seem to work with mlogit as it would with linear regression. Furthermore when I use the std command alone I can’t see the confidence intervals the standardized estimates. Can someone please point me to the right command or tell me how to produce the confidence intervals?

Second, I am unable to obtain standardized coefficient for my categorical predictor (smoke, yes/no).   The other predictors in my model are continuous. When I look at the output with  “standardized coefficients” there is no standardized coefficient for the categorical variable. I wonder what happened to the variable during the standardization?  Is a  categorical  accounted for or is it systematically dropped in this type of standardization?  It makes me wonder whether standardization is actually appropriate in this  case.

Perhaps someone tell me if the standardization process in the case of mlogit is similar to the one done in linear regression , i.e . for each variable the standardized version would be variable minus its  means divided by standard deviation? Or is it different?

I am using STATA11 and along with some of the options I've tried above, my code is mlogit group glucose age bmi smoke, base(0); Thank you in advance - any help is appreciated.

Lisa Staimez

________________________________

This e-mail message (including any attachments) is for the sole use of
the intended recipient(s) and may contain confidential and privileged
information. If the reader of this message is not the intended
recipient, you are hereby notified that any dissemination, distribution
or copying of this message (including any attachments) is strictly
prohibited.

If you have received this message in error, please contact
the sender by reply e-mail message and destroy all copies of the
original message (including attachments).

*
*   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/