0
$\begingroup$

I am performing a general-to-specific procedure. I want to check the joint significance of all the insignificant variables. I have a general model with all the variables, none removed, a model_x I have a second model with ONLY insignificant variables (and the dependent one), a model_y

The problem is, I have a time series object, an ARDL model with lags. In the general model, model_x, one of the variables has 4 lags. In the second model, that same variable has only 3 lags (because one of them was significant in the general model)

What I want to do now is that I want to compare both of those, model_x and model_y using the ANOVA function but I can't due to the different number of residuals.

'''

length(residuals(model_x))

[1] 115

length(residuals(model_y))

[1] 116 '''

I have no idea how to resolve this, help would be greatly appreciated

Improve this question
$\endgroup$

0

Reset to default

Your Answer

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

Browse other questions tagged or ask your own question.