st: standardised coefficients after lincom

[Christopher Hein <chein@ioe.ac.uk>]

Dear fellow users,
 
I would like to compute the  standardised coefficient after a linear combination test (lincom  command). In to complicate things all my variables I am using in lincom
 
lincom Xvar1+Xvar2+Xvar3
 
are interactions of a  continuous variable and a dummy. Now I know that some disciplines aren't  fond of standardised coefficients. This aside, I have found out that to  compute the standardised coefficient of an interaction variable (that I created manually, not by using xi:) STATA uses the  standard deviation of the corresponding un-interacted variable. To  compute the standardised coefficient STATA uses the standard error of  the estimate and NOT the standard deviation of the dummy variable.
 
So my question now is first: how to compute the standardised coefficient of any linear combination.
And secondly: what should I do in my case?
 
Note the normal formula to compute the standardised coefficient for variable X is:
beta=rawb_X*(sd_X/sd_Y),  where sd_X is the standard deviation of variable X and sd_Y the standard  deviation of the dependent variable.
 
Best,
Christopher F. Hein    
*
*   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/