Fixed Effect model - omitted because of collinearity

[Nicola Rossi]

Dear Stata Users,

I am applying a fixed effect reg model.

I want to study the relationship between Y and X1. Moreover, I also want to study how a moderator affects the relationship between Y and X1.
Thus, I include on the right-hand side of my model X1, the moderator, X1* the moderator, control variables.

Unfortunately, I obtain "Moderator omitted because of collinearity". So my output model is not complete.
It is probably due to the fact that the moderator is a context-level variable. It has much less variance than the other variables.
Have you ever had this kind of problem? Any suggestions would be really appreciated.


Thank you in advance for your support,
Regards

[Carlo Lazzaro]

Nicola:
acting on the FAQ is, as always, the best way of posting.
Interested listers can only guess-working what's the matter with your data and codes, as you do not provide any detail.
The most trivial reason is that your omitted predictor is a time-invariant one and, as expected, the -fe-machinery wiped it out.