In this lecture we will first introduce fixed and random effects one-way models for “balanced” panel data and learn 5 basic assumptions: (1)There are N cross sectional units each of which is observed T times in the data. (2)Short panel where time span is much smaller than the number of units. (3)Elements of error are identically and independently distributed. (4)Models are linear in parameters. (5)Independent variables may be correlated with error term. It is useful to think of 2SLS problems as ones where one focuses on one(for convenience, the first)structural equation of a system.