| Content | In this paper, we consider the instrumental variables estimation (the two-stage least squares estimator and the limited information maximum likelihood estimator) using weak instruments in a repeated measurements or a panel data model. We show that independently repeated cross-sectional data can reduce the asymptotic bias of the instrumental variables estimation when instruments are weakly correlated with endogenous variables. When the number of repeated measurements tends to infinity, we can achieve consistent instrumental variables estimation with weak instruments. |
