Mediation Analysis with the Mediator and Outcome Missing Not at Random
Open Access
- 26 June 2024
- journal article
- research article
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (550) , 794-804
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. 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.Keywords
All Related Versions
This publication has 43 references indexed in Scilit:
- UNPACKING COMPLEX MEDIATION MECHANISMS AND THEIR HETEROGENEITY BETWEEN SITES IN A JOB CORPS EVALUATIONJournal of Policy Analysis and Management, 2020
- Examining what factors mediate treatment effect in chronic low back pain: A mediation analysis of a Cognitive Functional Therapy clinical trialEuropean journal of pain, 2020
- Identifiability and estimation of causal mediation effects with missing dataStatistics in Medicine, 2017
- Identifiability of subgroup causal effects in randomized experiments with nonignorable missing covariatesStatistics in Medicine, 2013
- IDENTIFYING CAUSAL MECHANISMS (PRIMARILY) BASED ON INVERSE PROBABILITY WEIGHTINGJournal of Applied Econometrics, 2013
- Well-posedness of measurement error models for self-reported dataJournal of Econometrics, 2012
- Principles of Applied Statistics2011
- Test of a mediation model of perceived organizational supportJournal of Vocational Behavior, 2005
- Identification of graphical models for nonignorable nonresponse of binary outcomes in longitudinal studiesJournal of Multivariate Analysis, 2003
- On identifiability in models for incomplete binary dataStatistics & Probability Letters, 1999