In this paper we consider a class of structural econometric models in which the distribution of the endogenous variables can be defined as the solution of a fixed-point problem. We propose a two-stage estimator which avoids repeated solution of the fixed point problem. The idea behind our...