Refined GMM estimators for simultaneous equations models with network interactions

Abstract

The paper proposes a refinement of the generalized spatial two-stage and three-stage least squares estimators for simultaneous systems of equations with network interdependence, recently introduced in Drukker (Econom Theory 1-48, 2022). Specifically, we propose a refined weighting of the moment conditions underlying those estimators. Monte Carlo simulations document that the refined weighting potentially achieves non-trivial reductions in the root mean-squared errors for the network parameters of interest.