Publication Type

Journal Article

Version

submittedVersion

Publication Date

1-2024

Abstract

We consider a linear combination of jackknife Anderson-Rubin (AR) and orthogonalized Lagrangian multiplier (LM) tests for inference in IV regressions with many weak instruments and heteroskedasticity. We choose the weight in the linear combination based on a decision-theoretic rule that is adaptive to the identification strength. Under both weak and strong identifications, the proposed linear combination test controls asymptotic size and is admissible. Under strong identification, we further show that our linear combination test is the uniformly most powerful test against local alternatives among all tests that are constructed based on the jackknife AR and LM tests only and invariant to sign changes.

Keywords

Many instruments, power, size, weak identification

Discipline

Econometrics

Research Areas

Econometrics

Publication

Journal of Econometrics

Volume

238

Issue

2

First Page

1

Last Page

20

ISSN

0304-4076

Identifier

10.1016/j.jeconom.2023.105602

Publisher

Elsevier

Copyright Owner and License

Authors

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Additional URL

https://doi.org/10.1016/j.jeconom.2023.105602

Included in

Econometrics Commons

Share

COinS