st: Variable names and labels for interaction terms

["Nic" <3817629@hotmail.ca>]

Hello all,

I've figured out how to create all my (continuous x continuous) interaction 
variables at once but I'm trying to learn how to (1) automatically create 
variable labels for them, and (2) name the new variables based on truncated 
versions of the component variables' names.

Here is what I have so far:

foreach x of varlist s* {
foreach y of varlist f* {
generate i`x'_x_`y' =`x'*`y'
}
}

(1) I'd like each interaction term's variable label to be comprised of the 
component variables' labels. For example, a label that reads, "Physical 
Abuse x Self-Esteem".
I'm guessing this is possible since I can apply a single original variable 
label using the tips found here:
http://www.stata.com/support/faqs/data/variables.html
But I'm having trouble figuring out how to do something similar in the more 
complex scenario when two variables are involved. Any hints?

(2) I'd also like the interaction term variable names to be based on 
truncated versions of the component variables' names. For example, "is1f1" 
would be an ideal name for an interaction term created from 
"s1PhysicalAbuse" and "f1SelfEsteem". Currently my code creates variable 
names that are quite lengthy, for example, 
"is1PhysicalAbuse_x_f1SelfEsteem". It would save a lot of typing in later 
analyses if I were able to refer to a particular interaction by typing just 
the prefixes for the predictor and moderator variables. Again, any hints?

With thanks,
Nic 

*
*   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/