Rank Tests for Instrumental Variables Regression with Weak Instruments

52 Pages Posted: 31 Mar 2006

See all articles by Donald W. K. Andrews

Donald W. K. Andrews

Yale University - Cowles Foundation

Gustavo Soares

Yale University - Department of Economics

Date Written: March 2006

Abstract

This paper considers tests in an instrumental variables (IVs) regression model with IVs that may be weak. Tests that have near-optimal asymptotic power properties with Gaussian errors for weak and strong IVs have been determined in Andrews, Moreira, and Stock (2006a). In this paper, we seek tests that have near-optimal asymptotic power with Gaussian errors and improved power with non-Gaussian errors relative to existing tests. Tests with such properties are obtained by introducing rank tests that are analogous to the conditional likelihood ratio test of Moreira (2003). We also introduce a rank test that is analogous to the Lagrange multiplier test of Kleibergen (2002) and Moreira (2001).

Keywords: Asymptotically similar tests, Conditional likelihood ratio test, Instrumental variables regression, Lagrange multiplier test, Power of test, Rank tests, Thick-tailed distribution, Weak instruments

JEL Classification: C12, C30

Suggested Citation

Andrews, Donald W. K. and Soares, Gustavo, Rank Tests for Instrumental Variables Regression with Weak Instruments (March 2006). Cowles Foundation Discussion Paper No. 1564, Available at SSRN: https://ssrn.com/abstract=894330

Donald W. K. Andrews (Contact Author)

Yale University - Cowles Foundation ( email )

Box 208281
New Haven, CT 06520-8281
United States
203-432-3698 (Phone)
203-432-6167 (Fax)

Gustavo Soares

Yale University - Department of Economics ( email )

28 Hillhouse Ave
New Haven, CT 06520-8268
United States