Skip to main content

Rethinking Nonlinear Instrumental Variable Models through Prediction Validity

Publication ,  Journal Article
Li, C; Rudin, C; McCormick, TH
Published in: Journal of Machine Learning Research
January 1, 2022

Instrumental variables (IV) are widely used in the social and health sciences in situations where a researcher would like to measure a causal effect but cannot perform an experiment. For valid causal inference in an IV model, there must be external (exogenous) variation that (i) has a suffciently large impact on the variable of interest (called the relevance as- sumption) and where (ii) the only pathway through which the external variation impacts the outcome is via the variable of interest (called the exclusion restriction). For statistical inference, researchers must also make assumptions about the functional form of the relationship between the three variables. Current practice assumes (i) and (ii) are met, then postulates a functional form with limited input from the data. In this paper, we describe a framework that leverages machine learning to validate these typically unchecked but consequential assumptions in the IV framework, providing the researcher empirical evidence about the quality of the instrument given the data at hand. Central to the proposed approach is the idea of prediction validity. Prediction validity checks that error terms { which should be independent from the instrument { cannot be modeled with machine learning any better than a model that is identically zero. We use prediction validity to develop both one-stage and two-stage approaches for IV, and demonstrate their performance on an example relevant to climate change policy.

Duke Scholars

Published In

Journal of Machine Learning Research

EISSN

1533-7928

ISSN

1532-4435

Publication Date

January 1, 2022

Volume

23

Related Subject Headings

  • Artificial Intelligence & Image Processing
  • 4905 Statistics
  • 4611 Machine learning
  • 17 Psychology and Cognitive Sciences
  • 08 Information and Computing Sciences
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Li, C., Rudin, C., & McCormick, T. H. (2022). Rethinking Nonlinear Instrumental Variable Models through Prediction Validity. Journal of Machine Learning Research, 23.
Li, C., C. Rudin, and T. H. McCormick. “Rethinking Nonlinear Instrumental Variable Models through Prediction Validity.” Journal of Machine Learning Research 23 (January 1, 2022).
Li C, Rudin C, McCormick TH. Rethinking Nonlinear Instrumental Variable Models through Prediction Validity. Journal of Machine Learning Research. 2022 Jan 1;23.
Li, C., et al. “Rethinking Nonlinear Instrumental Variable Models through Prediction Validity.” Journal of Machine Learning Research, vol. 23, Jan. 2022.
Li C, Rudin C, McCormick TH. Rethinking Nonlinear Instrumental Variable Models through Prediction Validity. Journal of Machine Learning Research. 2022 Jan 1;23.

Published In

Journal of Machine Learning Research

EISSN

1533-7928

ISSN

1532-4435

Publication Date

January 1, 2022

Volume

23

Related Subject Headings

  • Artificial Intelligence & Image Processing
  • 4905 Statistics
  • 4611 Machine learning
  • 17 Psychology and Cognitive Sciences
  • 08 Information and Computing Sciences