Interaction term interpretation

[Michael Lee]

Hello,

In a model with interaction term like this:

y=a+b*x+c*z+d*x*z


If the coefficient of b in the model without interaction is negative, and the coefficient of d in model with interaction is positive how can I interpret this condition?


Thank you in advance
Best regards,
Michael

[Carlo Lazzaro]

Michael:
without any data sharing from your side, my only advice is to take a look at https://stats.oarc.ucla.edu/stata/se...actions-stata/ and related stuff.

[Maxence Morlet]

Are your regressors binary? Discrete? Continuous?

[Michael Lee]

Continuous, both of them