st: splines with interactions

[Dan MacNulty <macn0007@umn.edu>]

Dear Stata Users,
I'm modeling body growth using the -mkspline- command and I was looking 
for some advice on how to include interactions between the splines and 
gender. Specifically, if my model has 3 splines (age_1, age_2, age_3), 
is it necessary, from a statistical perspective, to include a gender 
interaction for the third spline (age_3*sex) even if only the gender 
interaction with the second spline (age_2*sex) is significant. In other 
words, does a model such as weight = age_1 + age_2 + age_3 + sex + 
age_2*sex, violate any assumptions? Likewise, if the model had a 
significant interaction with the first spline (age_1*sex) is it also 
necessary to include both age_2*sex and age_3*sex, even if these are not 
significant?

Any insights would be most welcome.

Sincerely,
Dan MacNulty
University of Minnesota
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/