interaction term with quadratic term?

[natasha greene]

1) Y= X1+ X1*X1+X2 + X1*X2 + (X2)*(X1*X1) i.e. adding not only the interaction with the low level X1 but also the interaction with the quadratic term X1*X1.
2) Y= X1+ X1*X1+X2 + X1*X2 i.e. just adding the interaction between X2 and the lower level of X1.

x1 is a continuous var whereas x2 is gender.

which is correct? or both are?

[Carlo Lazzaro]

Natasha:
welcome to this forum.
Due to the limited details you give in your post (for instance, we do not know your data), I find difficult to reply positively.
That said, the best advice is to avoid creating interaction by hand and exploit the wonderful capabilities of -fvvarlist- instead.

[natasha greene]

thanks Carlo Lazzaro !

Y is a dummy variable (school enrollment). X1 is a continous variable the child's heath HAZ and X2 is gender.
i want to see if there is any nonlinear relationship which there is, as i have seen with mcp graphs.
but i want to know if its necessary to include the quadratic term interacted with the gender or
not including wouldnt make any difference

[Carlo Lazzaro]

Natasha:
if you suspect a non-linear relationship between Y and HAZ (I do not understand whether X1 and HAZ are the same thing or not), you should try:

probit Y c.HAZ##c.HAZ i.gender