Graphical Models for Processing Missing Data

This article reviews recent advances in missing data research using graphical models to represent multivariate dependencies. We first examine the limitations of traditional frameworks from three different perspectives: transparency, estimability, and testability. We then show how procedures based on graphical models can overcome these limitations and provide meaningful performance guarantees even when data are missing not at random (MNAR). In particular, we identify conditions that guarantee consistent estimation in broad categories of missing data problems, and derive procedures for implementing this estimation. Finally, we derive testable implications for missing data models in both missing at random and MNAR categories.

Automated summary

This article reviews recent advances in missing data research using graphical models, addressing the limitations of traditional frameworks and providing meaningful performance guarantees, even in cases of missing data not at random. The study identifies conditions for consistent estimation and procedures for implementation, as well as deriving testable implications for missing data models.