Estimating a Linear Simultaneous Equation Model with Panel Data

In this paper we consider a class of the GMM estimators for a linear simultaneous equation model with panel data. The model is characterized by the presence of unobserved individual effects as part of structural disturbances and by the correlation of certain time-varying and time-invariant exogenous variables with the individual effects. A class of single-equation and system GMM estimators is proposed on the basis of a class of instruments as a unifying theme. Small-sample performance of this class of estimators is also investigated by a series of sampling experiments