Professor Jungbin Hwang has had his article Simple and Trustworthy Cluster-Robust GMM Inference accepted for publication in the Journal of Econometrics, a top field-journal in econometrics.
This paper develops a new asymptotic theory for GMM estimation and inference in the presence of clustered dependence. The key feature of the alternative asymptotic is that the number of clusters is regarded as “fixed” as the sample size increases. The paper shows that the Wald and t-tests in two-step GMM are asymptotically pivotal only if one recenters the estimated moment process in the clustered covariance estimator (CCE). Also, the J statistic, the trinity of two-step GMM statistics (QLR, LM, and Wald), and the t statistic can be modified to have an asymptotic standard F distribution or t distribution.
The paper also first suggests a finite-sample variance correction in the literature of cluster-robust methods and further improves the F and t approximations’ accuracy. The proposed tests in this paper are very appealing to practitioners because the test statistics are simple modifications of conventional GMM test statistics, and critical values are readily available from F and t tables without any extra simulations or resampling steps.
This sole authored article is one of two that Professor Hwang has recently had accepted for publication. Details of the other article may be found at:
Professor Hwang to publish “A Doubly Corrected Robust Variance Estimator for Linear GMM” in the Journal of Econometrics