A doubly corrected robust variance estimator for linear GMM

We propose a new finite sample corrected variance estimator for the linear generalized method of moments (GMM) including one-step, two-step, and iterated estimators. Our formula also corrects over-identification bias in estimation on top commonly used correction Windmeijer (2005), which from estimating efficient weight matrix, so is doubly corrected. An important feature proposed double that it automatically provides robustness to misspecification moment condition. In contrast, conventional are inconsistent under misspecification. That is, this paper convenient way obtain improved inference correct specification against at same time.