I am running FE and RE regressions and my Prof. suggested that I run unit-root tests. I have been trying to do that with R but am not able to find a library that will handle unbalanced panels with gaps.
My question: can the unit-root test be performed on a subset of the larger model, i.e. can I use a panel with as few as 3 columns: date, group, variable_of_interest?
Essentially I am trying to remove the gaps in my data by downsizing to the essential variables, but I am not sure how this will effect the results of the unit-root test.
