Given a large dataset with nested observations, how would you restructure it for a longitudinal analysis at SAS?
