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: IV Regression with Possibly Uncorrelated Instruments

Abstract: This paper proposes a closed-form linear IV estimator which allows endogenous covariates to be weakly correlated or un-correlated but mean-dependent on instruments. Identification rests on (1) a weak uncorrelatedness exclusion restriction and (2) a weak relevance condition where covariates are mean-dependent on instruments. The significant weakening of the relevance condition does not come at the cost of a stronger exclusion restriction. The estimator is root-n-consistent and asymptotically normal. Monte Carlo simulations show the estimator exploits unknown forms of both monotone and non-monotone identifying variation equally well, and it incurs less bias and size distortion relative to conventional IV methods when instruments are weak. An empirical example illustrates the practical usefulness of the estimator.
Subjects: Econometrics (econ.EM)
Cite as: arXiv:2103.09621 [econ.EM]
  (or arXiv:2103.09621v1 [econ.EM] for this version)

Submission history

From: Emmanuel Tsyawo [view email]
[v1] Wed, 17 Mar 2021 13:11:52 GMT (26kb)

Link back to: arXiv, form interface, contact.