Corpus ID: 88522281

Paradoxes in instrumental variable studies with missing data and one-sided noncompliance

@article{Kennedy2017ParadoxesII,
  title={Paradoxes in instrumental variable studies with missing data and one-sided noncompliance},
  author={Edward H. Kennedy and D. Small},
  journal={arXiv: Methodology},
  year={2017}
}
  • Edward H. Kennedy, D. Small
  • Published 2017
  • Mathematics
  • arXiv: Methodology
  • It is common in instrumental variable studies for instrument values to be missing, for example when the instrument is a genetic test in Mendelian randomization studies. In this paper we discuss two apparent paradoxes that arise in so-called single consent designs where there is one-sided noncompliance, i.e., where unencouraged units cannot access treatment. The first paradox is that, even under a missing completely at random assumption, a complete-case analysis is biased when knowledge of one… CONTINUE READING

    Figures from this paper.

    Efficient Nonparametric Causal Inference with Missing Exposure Information
    1

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 27 REFERENCES
    Instrumental Variables Estimation with Partially Missing Instruments
    20
    Identification of Treatment Effects on the Treated with One-Sided Non-Compliance
    21
    Instrumental variable methods for causal inference.
    242
    IDENTIFICATION AND ESTIMATION OF TREATMENT EFFECTS WITH A REGRESSION-DISCONTINUITY DESIGN
    2078
    Instruments for Causal Inference: An Epidemiologist's Dream?
    682
    Estimating Causal Effects of Treatments in Randomized and Nonrandomized Studies
    6078
    GMM with Multiple Missing Variables
    8