Interaction term's significance changes by including the intercept. How to decide if intercept is to be kept or not?

Question

I am working on a model (non-linear) of the form:

Should I have the intercept in the equation, as having one gives a different outcome (level of significance) to my interaction term 1, compared to not having the intercept?

Answer 1

You should almost never leave the intercept out of the model, unless you have a very good reason, and what you have given is not a good reason. That p-value for an interaction changes is not a good reason.

Have a look at

• When is it ok to remove the intercept in a linear regression model?
• The difference between with or without intercept model in logistic regression