st: Decomposition of coefficient, two groups

[Laurie Molina <molinalaurie@gmail.com>]

Dear Stata list,
Consider a population divided in two groups: group 1 and group 2. By
appending group 1 and group 2, we get the entire sample, which
corresponds to the entire population.
Im interest in the effect of an explanatory variable X on the outcome
variable Y.
When considering the entire population, X has a significant and high
effect on Y (coefficient has value of 0.61 and there are 3.7 million
observations).
When considering only group 1, X has a significant and very high
effect on Y (coefficient has value of 1.4 and there are 1 million
observations).
When considering only group 2, X has a significant and low effect on Y
(coefficient has value of .42  and there are 2.7 million
observations)..
The effect of X on Y, is statistically different between group 1 and
group 2. The coefficients by group are estimated introducing dummies
for group 1 and group 2 respectively, as well as interaction terms for
all explanatory variables.

I'm interesting in finding "what fraction of the variation in Y
associated with X, is explained by being a member of group 1".
Is there any "coefficient decomposition" that I can perform in stata
to answer such a question?
All suggestions and comments are very much appreciated.
*
*   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/