Abstract
SummaryLikelihood factors that can be disregarded for inference are termed ignorable. We demonstrate that close ties exist between ignorability and identification of causal effects by covariate adjustment. A graphical condition, stability, plays a role analogous to that of missingness at random, but is applicable to general longitudinal data. Our formulation of ignorability does not depend on any notion of missing data, so is appealing in situations where missing data may not actually exist. Several examples illustrate how stability may be assessed.
Highlights
We consider the analysis of longitudinal data
We provide an alternative characterization of ignorability for general longitudinal data that does not depend on any notion of missing data
We contend that missing data are not nearly so widespread as their prominence in the statistical literature would imply
Summary
We consider the analysis of longitudinal data. For each of a set of subjects, a sequence of observations is recorded, corresponding to the same property or feature of the subject evaluated at different times. For longitudinal data, this is only very rarely the case; more usually, missing data constitute a convenient and sometimes compelling fiction. We argue that this fiction is not needed. That crown-rump length does not exist at every point after conception. A healthy foetus has a shape and size that is complex and growing more or less continuously, but crown-rump length is not a one-dimensional slice of this highdimensional, continuous-time process; it is an external procedure subject to many influences apart from foetal size, including the skill of the sonographer, the resolution of the ultrasound, and the cooperation of the foetus.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
