This paper develops a specification test for instrument validity in the heterogeneous treatment effect model with a binary treatment and a discrete instrument. The strongest testable implication for instrument validity is given by the condition for nonnegativity of point-identifiable compliers' outcome densities. Our specification test infers this testable implication using a variance-weighted Kolmogorov–Smirnov test statistic. The test can be applied to both discrete and continuous outcome cases, and an extension of the test to settings with conditioning covariates is provided.
Authors
Research Associate University College London and Brown University
Toru is a Research Associate of the IFS, a Professor of Economics at UCL and an Associate Professor in the Department of Economics at Brown University
Journal article details
- DOI
- 10.3982/ECTA11974
- Publisher
- Wiley
- Issue
- Volume 83, Issue 5, September 2015
Suggested citation
Kitagawa, T. (2015). 'A Test for Instrument Validity' 83(5/2015)
More from IFS
Understand this issue
Empty defence spending promises are a shot in the dark
29 April 2024
Public investment: what you need to know
25 April 2024
The £600 billion problem awaiting the next government
25 April 2024
Policy analysis
Recent trends in and the outlook for health-related benefits
19 April 2024
4.2 million working-age people now claiming health-related benefits, could rise by 30% by the end of the decade
19 April 2024
Progression of nurses within the NHS
12 April 2024
Academic research
The employment and distributional impacts of nationwide minimum wage changes
10 April 2024
Willingness to pay for improved public education and public healthcare systems: the role of income mobility prospects
14 March 2024
Unfunded mandates and taxation
14 March 2024