2
$\begingroup$

I have a multiple linear regression model (OLS) that has a constant, 5 variables, and an interaction term between two of the dummy variables.

I need to perform the Durbin-Watson test for first-order correlation, and in order to find the critical values, I need to know the number of explanatory variables (not taking into account the constant).

Is the interaction term an explanatory variable in itself for these purposes (so I have 6 explanatory vars), or do I find the critical values for 5 explanatory vars?

Thanks!

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

The interaction counts as an explanatory variable. If there is a factor involved in the interaction, all of the levels after the first count (so an interaction between two factors can count as many terms in this situation).

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.