Hello,
I am sorry if this would sound like trivial questions (I am new to longitudinal data analysis), but I was not able to find a clear answer online.
I have an unbalanced panel with 100 countries over 24 years and I am using an autoregressive model.

There are two issues that I am not sure about:
1) I am using
Code:
xtfisher
to test for unit root, as I have an unbalanced panel. Would you have any other test to suggest?
2) Should I be concerned only about my dependent variable or also about my regressors. While my dependent variable is not a unit root, some of the regressors are.
Should i first difference my regressors or I should not worry to much about it. I have been looking in some replication dataset of published studies and most of them seem to test UR only for the dependent variable. I just wanted to be sure this was correct.

Thank you in advance for your reply