st: xtmixed with heteroskedastic errors?

[David Airey <david.airey@Vanderbilt.Edu>]

In regards to "Tricks of the trade: Getting the most out of xtmixed"  
by Roberto G. Gutierrez, at <http://econpapers.repec.org/paper/bocfsug08/ 
>, this means that the examples in West, Welch, and Galecki, Linear  
Mixed Models: A Practical Guide Using Statistical Software, where the  
authors say xtmixed cannot be used (e.g., page 85) because it doesn't  
allow residual errors to differ between groups, is wrong? Maybe an  
email should be sent to update the authors. I know they have a  
companion web site with the Stata code for the examples in the text.  
If the text is not updated, then at least the web site materials could  
be.
*
*   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/