Hello,

I'm working with an unbalanced panel data and want to sub-sample on a dummy variable which has no mising values throughout the full sample. Is it normal that the outcome contain many empty observations?

artur