2020/08/20 by Rami V. Tabri, Tabri, Rami V., Christopher D. Walker +1
Economics, Econometrics and Finance · Mathematics · #Econometrics (econ.EM) #Economic Policies and Impacts #FOS: Economics and business #Global trade and economics #Monetary Policy and Economic Impact #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.2008.09021
openalex publication_date 2020/08/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Inference in models where the parameter is defined by moment inequalities is\nof interest in many areas of economics. This paper develops a new method for\nimproving the performance of generalized moment selection (GMS) testing\nprocedures in finite-samples. The method modifies GMS tests by tilting the\nempirical distribution in its moment selection step by an amount that maximizes\nthe empirical likelihood subject to the restrictions of the null hypothesis. We\ncharacterize sets of population distributions on which a modified GMS test is\n(i) asymptotically equivalent to its non-modified version to first-order, and\n(ii) superior to its non-modified version according to local power when the\nsample size is large enough. An important feature of the proposed modification\nis that it remains computationally feasible even when the number of moment\ninequalities is large. We report simulation results that show the modified\ntests control size well, and have markedly improved local power over their\nnon-modified counterparts.\n