We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

econ.EM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Economics > Econometrics

Title: Testing Finite Moment Conditions for the Consistency and the Root-N Asymptotic Normality of the GMM and M Estimators

Abstract: Common approaches to inference for structural and reduced-form parameters in empirical economic analysis are based on the consistency and the root-n asymptotic normality of the GMM and M estimators. The canonical consistency (respectively, root-n asymptotic normality) for these classes of estimators requires at least the first (respectively, second) moment of the score to be finite. In this article, we present a method of testing these conditions for the consistency and the root-n asymptotic normality of the GMM and M estimators. The proposed test controls size nearly uniformly over the set of data generating processes that are compatible with the null hypothesis. Simulation studies support this theoretical result. Applying the proposed test to the market share data from the Dominick's Finer Foods retail chain, we find that a common \textit{ad hoc} procedure to deal with zero market shares in analysis of differentiated products markets results in a failure to satisfy the conditions for both the consistency and the root-n asymptotic normality.
Subjects: Econometrics (econ.EM)
Cite as: arXiv:2006.02541 [econ.EM]
  (or arXiv:2006.02541v3 [econ.EM] for this version)

Submission history

From: Yulong Wang [view email]
[v1] Wed, 3 Jun 2020 21:32:12 GMT (46kb,D)
[v2] Tue, 30 Jun 2020 00:27:42 GMT (174kb,D)
[v3] Wed, 2 Sep 2020 19:58:26 GMT (176kb,D)

Link back to: arXiv, form interface, contact.