Re: [Gretl-users] sub-sampling panel data
artur.bala schrieb:
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?
not sure I understand -- do you mean there are missing values in the
other variables? Maybe the existence of missings is in some sense
correlated with the dummy variable you're using? Or do you mean the
missings suddenly appear after the subsampling and weren't there before?
cheers,
sven