Mediation analysis with the mediator and outcome missing not at random
Shuozhi Zuo, Debashis Ghosh, Peng Ding, and Fan Yang

TL;DR
This paper investigates the identifiability of direct and indirect effects in mediation analysis when both mediator and outcome are missing not at random, proposing methods under certain assumptions and demonstrating their performance through simulations and real data.
Contribution
It introduces conditions under which mediation effects are identifiable despite missing not at random data, and evaluates inference methods with simulations and a real case study.
Findings
Identifiability of effects depends on specific mechanisms for missingness.
Proposed methods perform well in simulation studies.
Application to National Job Corps Study illustrates practical utility.
Abstract
Mediation analysis is widely used for investigating direct and indirect causal pathways through which an effect arises. However, many mediation analysis studies are challenged by missingness in the mediator and outcome. In general, when the mediator and outcome are missing not at random, the direct and indirect effects are not identifiable without further assumptions. In this work, we study the identifiability of the direct and indirect effects under some interpretable mechanisms that allow for missing not at random in the mediator and outcome. We evaluate the performance of statistical inference under those mechanisms through simulation studies and illustrate the proposed methods via the National Job Corps Study.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Causal Inference Techniques
