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


Current browse context:


Change to browse by:

References & Citations


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

Economics > Econometrics

Title: Inference on Functionals of Set-Identified Parameters Defined by Convex Moments

Abstract: This paper considers uniformly valid (over a class of data generating processes) inference for convex functionals of partially identified parameters in cases where the identified set is defined by convex (in the parameter) moment inequalities. We propose a bootstrap procedure for constructing uniformly valid confidence sets for a convex functional of a partially identified parameter. The proposed method amounts to bootstrapping the value functions of a convex optimization problem, and subsumes subvector inference as a special case. Unlike other proposed subvector inference procedures, our procedure does not require the researcher to repeatedly invert a hypothesis test, and is extremely computationally efficient. Finally, we discuss sufficient conditions on the moment functions to ensure uniform validity, and show that our assumptions are in fact satisfied by a large class of data generating processes.
Subjects: Econometrics (econ.EM)
Cite as: arXiv:1810.03180 [econ.EM]
  (or arXiv:1810.03180v5 [econ.EM] for this version)

Submission history

From: Thomas Russell [view email]
[v1] Sun, 7 Oct 2018 17:03:14 GMT (59kb,D)
[v2] Wed, 2 Jan 2019 22:21:05 GMT (59kb,D)
[v3] Mon, 27 May 2019 21:46:18 GMT (66kb,D)
[v4] Sun, 9 Jun 2019 12:08:41 GMT (53kb,D)
[v5] Wed, 26 Jun 2019 13:01:33 GMT (53kb,D)
[v6] Sat, 17 Aug 2019 19:01:54 GMT (40kb,D)
[v7] Sun, 8 Sep 2019 18:14:17 GMT (42kb,D)

Link back to: arXiv, form interface, contact.