Identification with Imperfect Instruments
43 Pages Posted: 27 Oct 2008 Last revised: 11 Dec 2022
Date Written: October 2008
Abstract
Dealing with endogenous regressors is a central challenge of applied research. The standard solution is to use instrumental variables that are assumed to be uncorrelated with unobservables. We instead assume (i) the correlation between the instrument and the error term has the same sign as the correlation between the endogenous regressor and the error term, and (ii) that the instrument is less correlated with the error term than is the endogenous regressor. Using these assumptions, we derive analytic bounds for the parameters. We demonstrate the method in two applications.
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
Instrumental Variables Regression with Weak Instruments
By Douglas Staiger and James H. Stock
-
Testing for Weak Instruments in Linear IV Regression
By James H. Stock and Motohiro Yogo
-
Testing for Weak Instruments in Linear IV Regression
By James H. Stock and Motohiro Yogo
-
A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments
By James H. Stock, Jonathan H. Wright, ...
-
Instrument Relevance in Multivariate Linear Models: A Simple Measure
By John Shea
-
By Charles R. Nelson and Richard Startz
-
Some Further Results on the Exact Small Sample Properties of the Instrumental Variable Estimator
By Charles R. Nelson and Richard Startz
-
A New Specification Test for the Validity of Instrumental Variables
By Jinyong Hahn and Jerry A. Hausman
-
Consistent Estimation with a Large Number of Weak Instruments
By John C. Chao and Norman R. Swanson
-
Does Head Start Make a Difference?
By Janet Currie and Duncan Thomas